https://visone.ethz.ch/wiki/api.php?action=feedcontributions&feedformat=atom&user=Madervisone manual - User contributions [en]2026-09-10T18:58:58ZUser contributionsMediaWiki 1.43.3https://visone.ethz.ch/wiki/index.php?title=Complement_Graph&diff=1607Complement Graph2015-06-08T10:18:26Z<p>Mader: Created page with "This function creates the complement graph corresponding to the current network. The resulting complement graph contains a link between two nodes exactly if there is no link in t..."</p>
<hr />
<div>This function creates the complement graph corresponding to the current network. The resulting complement graph contains a link between two nodes exactly if there is no link in the original graph and vice versa.</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=675Visualization tab2011-04-15T07:35:22Z<p>Mader: /* geometry */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance of force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
<!-- See the link routing [[link routing|concept page]] for details on available options. --><br />
<br />
=== label placement ===<br />
<br />
rearranges label positions to avoid overlap with nodes and links.<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for both nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
A mapping for node and link attributes to the color of nodes, node borders and links, suitable for both numerical and categorial attributes. <br />
For numerical attributes, values can be mapped by means of <br />
* interpolation in the RGB color space between a given color for the minimum value and a given color for the maximum value.<br />
* saturation of a given color.<br />
* brightness of a given color.<br />
Categorial attributes can be mapped by specifying a color table.<br />
<br />
==== size ====<br />
<br />
A mapping for node and link attributes to node area, width or height, or link width. <br />
This mapping is only allowed for numerical attributes.<br />
<br />
==== label ====<br />
<br />
A mapping for node and link attributes to properties of the corresponding labels. <br />
Available options are:<br />
* plain mapping of the attribute value to label texts.<br />
* mapping of values given by the attribute to label colors (in the same way as color-mapping for nodes and links).<br />
* mapping of numerical attribute values to font-sizes of labels.<br />
<br />
==== coordinates ====<br />
<br />
A mapping for numerical node attributes to coordinates in either x- or y-direction. <br />
Attribute values can be applied directly, or interploated to fit into the dimensions of the network's current diagram. <br />
Horizontal or vertical lines drawn in the background indicate corresponding attribute values.<br />
<br />
==== z-layer ====<br />
<br />
A mapping for numerical node attributes to the order in which nodes are drawn in the diagram. <br />
In case of overlap, nodes with a lower value will be beneath nodes with a higher value.<br />
<br />
== geometry ==<br />
<br />
The geometry category provides mechanisms to alter the layout of a network by means of purely geometric operations, so called affine transformations.<br />
<br />
=== affine transformations ===<br />
<br />
Affine transformations provided in visone are translation, rotation, scaling and reflection. <br />
<br />
=== procrustes analysis ===<br />
<br />
Networks in a network collection are automatically translated, rotated, and scaled to match the cofiguration of a reference network or the previous network in the collection as good as possible.</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=674Visualization tab2011-04-13T15:58:03Z<p>Mader: /* geometry */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance of force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
<!-- See the link routing [[link routing|concept page]] for details on available options. --><br />
<br />
=== label placement ===<br />
<br />
rearranges label positions to avoid overlap with nodes and links.<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for both nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
A mapping for node and link attributes to the color of nodes, node borders and links, suitable for both numerical and categorial attributes. <br />
For numerical attributes, values can be mapped by means of <br />
* interpolation in the RGB color space between a given color for the minimum value and a given color for the maximum value.<br />
* saturation of a given color.<br />
* brightness of a given color.<br />
Categorial attributes can be mapped by specifying a color table.<br />
<br />
==== size ====<br />
<br />
A mapping for node and link attributes to node area, width or height, or link width. <br />
This mapping is only allowed for numerical attributes.<br />
<br />
==== label ====<br />
<br />
A mapping for node and link attributes to properties of the corresponding labels. <br />
Available options are:<br />
* plain mapping of the attribute value to label texts.<br />
* mapping of values given by the attribute to label colors (in the same way as color-mapping for nodes and links).<br />
* mapping of numerical attribute values to font-sizes of labels.<br />
<br />
==== coordinates ====<br />
<br />
A mapping for numerical node attributes to coordinates in either x- or y-direction. <br />
Attribute values can be applied directly, or interploated to fit into the dimensions of the network's current diagram. <br />
Horizontal or vertical lines drawn in the background indicate corresponding attribute values.<br />
<br />
==== z-layer ====<br />
<br />
A mapping for numerical node attributes to the order in which nodes are drawn in the diagram. <br />
In case of overlap, nodes with a lower value will be beneath nodes with a higher value.<br />
<br />
== geometry ==<br />
<br />
The geometry category provides means to alter the layout of a network by means of purely geometric operations, so called affine transformations.<br />
<br />
=== affine transformations ===<br />
<br />
Affine transformations provided in visone are translation, rotation, scaling and reflection. <br />
<br />
=== procrustes analysis ===<br />
<br />
Networks in a network collection are automatically translated, rotated, and scaled to match the cofiguration of a reference network or the previous network in the collection as good as possible.</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=673Visualization tab2011-04-13T15:18:37Z<p>Mader: /* spring embedder */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance of force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
<!-- See the link routing [[link routing|concept page]] for details on available options. --><br />
<br />
=== label placement ===<br />
<br />
rearranges label positions to avoid overlap with nodes and links.<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for both nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
A mapping for node and link attributes to the color of nodes, node borders and links, suitable for both numerical and categorial attributes. <br />
For numerical attributes, values can be mapped by means of <br />
* interpolation in the RGB color space between a given color for the minimum value and a given color for the maximum value.<br />
* saturation of a given color.<br />
* brightness of a given color.<br />
Categorial attributes can be mapped by specifying a color table.<br />
<br />
==== size ====<br />
<br />
A mapping for node and link attributes to node area, width or height, or link width. <br />
This mapping is only allowed for numerical attributes.<br />
<br />
==== label ====<br />
<br />
A mapping for node and link attributes to properties of the corresponding labels. <br />
Available options are:<br />
* plain mapping of the attribute value to label texts.<br />
* mapping of values given by the attribute to label colors (in the same way as color-mapping for nodes and links).<br />
* mapping of numerical attribute values to font-sizes of labels.<br />
<br />
==== coordinates ====<br />
<br />
A mapping for numerical node attributes to coordinates in either x- or y-direction. <br />
Attribute values can be applied directly, or interploated to fit into the dimensions of the network's current diagram. <br />
Horizontal or vertical lines drawn in the background indicate corresponding attribute values.<br />
<br />
==== z-layer ====<br />
<br />
A mapping for numerical node attributes to the order in which nodes are drawn in the diagram. <br />
In case of overlap, nodes with a lower value will be beneath nodes with a higher value.<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=672Visualization tab2011-04-13T15:15:59Z<p>Mader: /* mapping */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
<!-- See the link routing [[link routing|concept page]] for details on available options. --><br />
<br />
=== label placement ===<br />
<br />
rearranges label positions to avoid overlap with nodes and links.<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for both nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
A mapping for node and link attributes to the color of nodes, node borders and links, suitable for both numerical and categorial attributes. <br />
For numerical attributes, values can be mapped by means of <br />
* interpolation in the RGB color space between a given color for the minimum value and a given color for the maximum value.<br />
* saturation of a given color.<br />
* brightness of a given color.<br />
Categorial attributes can be mapped by specifying a color table.<br />
<br />
==== size ====<br />
<br />
A mapping for node and link attributes to node area, width or height, or link width. <br />
This mapping is only allowed for numerical attributes.<br />
<br />
==== label ====<br />
<br />
A mapping for node and link attributes to properties of the corresponding labels. <br />
Available options are:<br />
* plain mapping of the attribute value to label texts.<br />
* mapping of values given by the attribute to label colors (in the same way as color-mapping for nodes and links).<br />
* mapping of numerical attribute values to font-sizes of labels.<br />
<br />
==== coordinates ====<br />
<br />
A mapping for numerical node attributes to coordinates in either x- or y-direction. <br />
Attribute values can be applied directly, or interploated to fit into the dimensions of the network's current diagram. <br />
Horizontal or vertical lines drawn in the background indicate corresponding attribute values.<br />
<br />
==== z-layer ====<br />
<br />
A mapping for numerical node attributes to the order in which nodes are drawn in the diagram. <br />
In case of overlap, nodes with a lower value will be beneath nodes with a higher value.<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=671Visualization tab2011-04-13T15:14:03Z<p>Mader: /* label placement */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
<!-- See the link routing [[link routing|concept page]] for details on available options. --><br />
<br />
=== label placement ===<br />
<br />
rearranges label positions to avoid overlap with nodes and links.<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for both nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
A mapping for node and link attributes to the color of nodes, node borders and links, suitable for both numerical and categorial attributes. <br />
For numerical attributes, values can be mapped by means of <br />
* interpolation in the RGB color space between a given color for the minimum value and a given color for the maximum value.<br />
* saturation of a given color.<br />
* brightness of a given color.<br />
Categorial attributes can be mapped by specifying a color table.<br />
<br />
==== size ====<br />
<br />
A mapping for node and link attributes to node area, width or height, or link width. <br />
This mapping is only allowed for numerical attributes.<br />
<br />
==== label ====<br />
<br />
A mapping for node and link attributes to properties of the corresponding labels. <br />
Available options are:<br />
* plain mapping of the attribute value to label texts.<br />
* mapping of values given by the attribute to label colors (in the same way as color-mapping for nodes and links).<br />
* mapping of numerical attribute values to font-sizes of labels.<br />
<br />
==== coordinates ====<br />
<br />
A mapping for numerical node attributes to coordinates in either x- or y-direction. <br />
Attribute values can be applied directly, or interploated to fit into the dimensions of the network's current diagram. <br />
Horizontal or vertical lines drawn in the background indicate corresponding attribute values.<br />
<br />
<br />
==== z-layer ====<br />
<br />
A mapping for numerical node attributes to the order in which nodes are drawn in the diagram. <br />
In case of overlap, nodes with a lower value will be beneath nodes with a higher value.<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=670Visualization tab2011-04-13T15:08:26Z<p>Mader: /* mapping */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
<!-- See the link routing [[link routing|concept page]] for details on available options. --><br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for both nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
A mapping for node and link attributes to the color of nodes, node borders and links, suitable for both numerical and categorial attributes. <br />
For numerical attributes, values can be mapped by means of <br />
* interpolation in the RGB color space between a given color for the minimum value and a given color for the maximum value.<br />
* saturation of a given color.<br />
* brightness of a given color.<br />
Categorial attributes can be mapped by specifying a color table.<br />
<br />
==== size ====<br />
<br />
A mapping for node and link attributes to node area, width or height, or link width. <br />
This mapping is only allowed for numerical attributes.<br />
<br />
==== label ====<br />
<br />
A mapping for node and link attributes to properties of the corresponding labels. <br />
Available options are:<br />
* plain mapping of the attribute value to label texts.<br />
* mapping of values given by the attribute to label colors (in the same way as color-mapping for nodes and links).<br />
* mapping of numerical attribute values to font-sizes of labels.<br />
<br />
==== coordinates ====<br />
<br />
A mapping for numerical node attributes to coordinates in either x- or y-direction. <br />
Attribute values can be applied directly, or interploated to fit into the dimensions of the network's current diagram. <br />
Horizontal or vertical lines drawn in the background indicate corresponding attribute values.<br />
<br />
<br />
==== z-layer ====<br />
<br />
A mapping for numerical node attributes to the order in which nodes are drawn in the diagram. <br />
In case of overlap, nodes with a lower value will be beneath nodes with a higher value.<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=669Visualization tab2011-04-13T14:42:30Z<p>Mader: /* link routing */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
<!-- See the link routing [[link routing|concept page]] for details on available options. --><br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for aboth nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=668Visualization tab2011-04-13T14:42:06Z<p>Mader: /* node layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
<!-- For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization. --><br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
<!-- See the metric MDS [[metric MDS|concept page]] for more technical details. --><br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
<!-- See the centrality layout [[centrality layout|concept page]] for details on available options. --><br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
<!-- See the status layout [[status layout|concept page]] for more details. --><br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
<!-- See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options. --><br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
<!-- See the spring embedder [[spring embedder|concept page]] for further details. --><br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
<!-- See the spectral layout [[spectral layout|concept page]] for more details. --><br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
<!-- See the circular layout [[circular layout|concept page]] for details. --><br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
See the link routing [[link routing|concept page]] for details on available options.<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for aboth nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=667Visualization tab2011-04-13T14:40:42Z<p>Mader: /* mapping */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
See the status layout [[status layout|concept page]] for more details.<br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options.<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
See the spring embedder [[spring embedder|concept page]] for further details.<br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
See the spectral layout [[spectral layout|concept page]] for more details.<br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
See the circular layout [[circular layout|concept page]] for details.<br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
See the link routing [[link routing|concept page]] for details on available options.<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color]], [[#size|size]], and [[#label|labels]]. These are available for aboth nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=666Visualization tab2011-04-13T14:40:17Z<p>Mader: /* mapping */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
See the status layout [[status layout|concept page]] for more details.<br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options.<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
See the spring embedder [[spring embedder|concept page]] for further details.<br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
See the spectral layout [[spectral layout|concept page]] for more details.<br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
See the circular layout [[circular layout|concept page]] for details.<br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
See the link routing [[link routing|concept page]] for details on available options.<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
Methods in the mapping category are all used to map attribute values to visual attributes of the network diagram. <br />
Graphical attributes are [[#color|color], [[#size|size]], and [[#label|labels]]. These are available for aboth nodes and links.<br />
Mappings only available for nodes are the mappings to the geometrical attributes [[#coordinates|coordinates]], and the [[#z-layer|order]] in wich nodes are rendered.<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=665Visualization tab2011-04-13T12:31:27Z<p>Mader: /* dynamic layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
See the status layout [[status layout|concept page]] for more details.<br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides [[animation]] of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options.<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
See the spring embedder [[spring embedder|concept page]] for further details.<br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
See the spectral layout [[spectral layout|concept page]] for more details.<br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
See the circular layout [[circular layout|concept page]] for details.<br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
See the link routing [[link routing|concept page]] for details on available options.<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=664Visualization tab2011-04-13T12:27:43Z<p>Mader: /* link routing */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
See the status layout [[status layout|concept page]] for more details.<br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides animation of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options.<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
See the spring embedder [[spring embedder|concept page]] for further details.<br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
See the spectral layout [[spectral layout|concept page]] for more details.<br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
See the circular layout [[circular layout|concept page]] for details.<br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
Links are re-routed to avoid overlap between nodes and links, by introducing or moving bend-points of links. <br />
<br />
See the link routing [[link routing|concept page]] for details on available options.<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=663Visualization tab2011-04-13T12:12:44Z<p>Mader: /* node layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
See the status layout [[status layout|concept page]] for more details.<br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides animation of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options.<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
The spring embedder is the most commonly known general-purpose layout technique (however, we generally advise to use [[#stress minimization|stress minimization]] due to better scaling and quality). <br />
The available method is an instance force-directed methods in which a network is likened to a physical system of repelling objects (the nodes) and springs of a given length (the links) binding adjacent nodes together. <br />
Nodes are iteratively repositioned based on the forces exerted on them, so that the system moves toward a force equilibrium.<br />
<br />
See the spring embedder [[spring embedder|concept page]] for further details.<br />
<br />
==== spectral layout ====<br />
<br />
Spectral layout computes positions based on Eigenvectors of a networks adjacency matrix or the corresponding Laplacian.<br />
<br />
See the spectral layout [[spectral layout|concept page]] for more details.<br />
<br />
==== circular layout ====<br />
<br />
Circular layout arranges nodes on one or more circles, based on the connectedness in the network.<br />
<br />
See the circular layout [[circular layout|concept page]] for details.<br />
<br />
==== random layout ====<br />
<br />
Random layout assigns a random position for each node.<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=662Visualization tab2011-04-13T11:42:37Z<p>Mader: /* node layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
See the status layout [[status layout|concept page]] for more details.<br />
<br />
==== dynamic layout ====<br />
<br />
Dynamic layout refers to techniques for obtaining coherent positions of nodes for several networks, usually embodying a time-series of a network. <br />
The goal is to convey the evolution of the network, by easing comparison between a current layout and the preceeding one. <br />
In visone, three methods are currently offered to obtain a dynamic layout, all based on [[#stress minimization|stress minimization]].<br />
One approach calculates one single layout for an aggregate of all individual networks, and applies these positions, i.e., all nodes strictly maintain their position throughout the sequence. <br />
The other two approaches trade off this perfect stability for better readability of individual networks. <br />
<br />
Note that dynamic layout can only be applied to [[network collection|network collections]]. <br />
Also, visone provides animation of the sequence via the [[File:Animation.png|link=animation]] icon in the [[GUI#toolbar|toolbar]]. <br />
<br />
See the dynamic layout [[dynamic layout|concept page]] for more information on technical details and available options.<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
This method differs from standard [[#stress minimization|stress minization]] in the choice of input distances. <br />
Usually, shortest-path distances are considered. <br />
Here, any numerical dyad attribute may be choosen as input.<br />
The method then tries to arrange nodes, such that those distances are matched as well as possible in the layout. <br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=661Visualization tab2011-04-12T14:54:21Z<p>Mader: /* node layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
Status layout, like [[#centrality layout|centrality layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on horizontal lines, where nodes with higher value are closer to the top, and nodes with lower value are closer to the bottom of the drawing. <br />
The algorithm tries to avoid link crossings. <br />
Note also, that bend-points for the links are introduced to increase readability. <br />
Optionally, the status niveaus (i.e., horizontal lines) corresponding to attribute values can be shown in the background of the visualization.<br />
<br />
See the status layout [[status layout|concept page]] for more details.<br />
<br />
==== dynamic layout ====<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=660Visualization tab2011-04-12T14:37:10Z<p>Mader: /* node layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
Metric MDS, also referred to as ''classical scaling'' is the original, spectral-decomposition variant of multidimensional scaling. <br />
As with [[#stress minimization|stress minimization]], the goal is to represent shortest-path distances as well as possible. <br />
In contrast to stress minimization all distances are treated equally. <br />
Thus, this approach generally yields good representation of large distance, but poor representation of shorter distances, affecting layout quality. <br />
However, since metric MDS produces a unique solution, it is suited well to serve as initialization for stress minimization.<br />
<br />
See the metric MDS [[metric MDS|concept page]] for more technical details.<br />
<br />
==== centrality layout ====<br />
<br />
Centrality layout, like [[#status layout|status layout]] is used to obtain a layout that represents values of a given numerical nodal attribute, e.g. a centrality index of nodes. <br />
Nodes with the same attribute value are arranged on concentric circles, where nodes with higher value are closer to the center, and nodes with lower value are in the periphery.<br />
At the same time, the algorithm tries to reduce link crossings as much as possible. <br />
Circumferences corresponding to regular intervals with respect to the attribute values are shown in the background of the visualization, to increase legibility of each node's corresponding value.<br />
<br />
See the centrality layout [[centrality layout|concept page]] for details on available options.<br />
<br />
==== status layout ====<br />
<br />
==== dynamic layout ====<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=659Visualization tab2011-04-12T14:11:16Z<p>Mader: /* stress minimization */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization, an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS), is our preferred method to obtain a general-purpose layout for networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Note that the outcome of stress minimization is dependent on the current layout of the network.<br />
We suggest to compute a [[#metric MDS|metric MDS]] layout first to obtain good results.<br />
Also note that computing a layout via the [[Quick_layout|quick layout]] button corresponds to this procedure, i.e., applying stress minimization to a metric MDS layout.<br />
<br />
For more details on the options available in visone, see the [[stress minimization|concept page]] for stress minimization.<br />
<br />
==== metric MDS ====<br />
<br />
==== centrality layout ====<br />
<br />
==== status layout ====<br />
<br />
==== dynamic layout ====<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=658Visualization tab2011-04-12T13:33:11Z<p>Mader: /* node layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* given a [[#dynamic layout|sequence of networks]], the layout should ease comparison with respect to the layout of the previous network in the sequence.<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization is our preferred method to obtain a layout for general networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Stress minimization is an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS).<br />
<br />
==== metric MDS ====<br />
<br />
==== centrality layout ====<br />
<br />
==== status layout ====<br />
<br />
==== dynamic layout ====<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=657Visualization tab2011-04-12T13:31:19Z<p>Mader: /* layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|computing node positions]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus, most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
* links should have more or less the same length.<br />
* nodes should be distributed well over the drawing area.<br />
* the number of meaningless link crossings should be kept small.<br />
* structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* the layout should ease comparison with respect to the previous layout, given a [[#dynamic layout|sequence of networks]].<br />
* the layout should reveal [[#spectral layout|specific structural properties]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
Stress minimization is our preferred method to obtain a layout for general networks.<br />
The main idea is to compute a layout such that graph-theoretic distances (i.e., shortest-path lengths) between nodes are represented as good as possible, where more weight is placed on representation error with respect to shorter distances than larger ones. <br />
The method usually meets the general criteria mentioned above, and yields better results than [[#spring embedder|spring embedders]] in most cases.<br />
<br />
Stress minimization is an instance of a family of dimension-reduction techniques referred to as ''multidimensional scaling'' (MDS).<br />
<br />
==== metric MDS ====<br />
<br />
==== centrality layout ====<br />
<br />
==== status layout ====<br />
<br />
==== dynamic layout ====<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=656Visualization tab2011-04-12T12:47:42Z<p>Mader: /* layout */</p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
''Layout'' refers to the task of obtaining positions for the elements of a network visualization, where [[#node layout|arranging positions of nodes]] is of primary interest. <br />
Other tasks are [[#link routing|(re-)routing the links]] of a visualization, e.g. to avoid overlap between link and node representations, or to automatically [[#label placement|arrange label positions]] for better readability.<br />
<br />
<br />
=== node layout ===<br />
<br />
The methods in this section deal with the computation of node positions for one or more networks. Generally, nodes are considered to be geometric points (or objects that are described by a single point), and links are represented as straight lines between their incident nodes. <br />
Thus most methods produce so called ''straight-line drawings'' (also referred to as ''node-link diagrams'' or ''sociograms''). <br />
<br />
There are several general objectives that most methods try to optimize, such as:<br />
;uniform link length<br />
:links should have more or less the same length.<br />
;node distribution<br />
:nodes should be distributed well over the drawing area.<br />
;minimal crossings<br />
:the number of meaningless link crossings should be kept small.<br />
;symmetries<br />
:structural symmetries in the network should be represented well.<br />
<br />
Additionally, some methods are constrained by additional or different objectives:<br />
* node placement is restricted with respect to a given scalar node attribute, e.g., such that nodes lie on [[#centrality layout|concentric circles]] or [[#status layout| verical layers]] corresponding to the attributes values.<br />
* the layout should ease comparison with respect to the previous layout, given a [[#dynamic layout|sequence of networks]].<br />
* the layout should reveal the [[#spectral layout|properties of a network's spectrum]].<br />
<br />
<br />
<br />
==== stress minimization ====<br />
<br />
==== metric MDS ====<br />
<br />
==== centrality layout ====<br />
<br />
==== status layout ====<br />
<br />
==== dynamic layout ====<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Visualization_tab&diff=655Visualization tab2011-04-12T12:02:32Z<p>Mader: </p>
<hr />
<div>Visualization algorithms change the graphical appearance of the network; they are accesible via the visualization tab. Basic illustrations of how to layout networks or display attribute values are provided in the two trails on [[Visualization_and_analysis_(trail)|''visualization and analysis'']] and [[Managing_attributes_(trail)|''advanced attribute management'']].<br />
<br />
visone distinguishes between three major '''visualization categories'''<br />
* ''layout'' to recompute the positions (coordinates) or nodes, links-bends, or labels to optimize readability or other specified layout criteria;<br />
* ''mapping'' to specify how attribute values (such as node centrality, tie strength, or class membership) are encoded in grapical variables (such as size, width, or color);<br />
* ''geometry'' to apply geometric transformations such as rotation, reflection, or scaling to the network or parts of the network; <br />
<br />
== layout ==<br />
<br />
=== node layout ===<br />
<br />
==== stress minimization ====<br />
<br />
==== metric MDS ====<br />
<br />
==== centrality layout ====<br />
<br />
==== status layout ====<br />
<br />
==== dynamic layout ====<br />
<br />
==== stress minimization (dyad attributes) ====<br />
<br />
==== spring embedder ====<br />
<br />
==== spectral layout ====<br />
<br />
==== circular layout ====<br />
<br />
==== random layout ====<br />
<br />
<br />
=== link routing ===<br />
<br />
...<br />
<br />
<br />
=== label placement ===<br />
<br />
...<br />
<br />
<br />
== mapping ==<br />
<br />
==== color ====<br />
<br />
==== size ====<br />
<br />
==== label ====<br />
<br />
==== coordinates ====<br />
<br />
==== z-layer ====<br />
<br />
<br />
== geometry ==<br />
<br />
=== affine transformations ===<br />
<br />
<br />
...<br />
<br />
=== procrustes analysis ===<br />
<br />
...</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Attribute_manager&diff=368Attribute manager2011-02-01T07:45:38Z<p>Mader: </p>
<hr />
<div>The attribute manager is started by clicking on its icon [[File:Attribute_manager.png|link=attribute_manager]] in visone's toolbar or pressing the attribute manager button in the [[selection_tab|selection tab]].<br />
<br />
As opposed to the attributes tab of the [[node_properties_dialog#attributes|node properties dialog]] and [[link_properties_dialog#attributes|link properties dialog]], that only allow to change the value of already existing attributes manually, the attribute manager provides a variety of possibilities related to the creation, configuration and automatical modification of attributes.<br />
<br />
== categories ==<br />
<br />
There are 4 times 4 (2) selectable categories within the attribute manager, e.g. node attribute configuration:<br />
<br />
[[File:Attribute_manager2.png]]<br />
<br />
The four radio buttons in the top line are used to select the kind of attribute one wants to manage, namely<br />
* node attributes<br />
* link attributes<br />
* dyad attributes<br />
* graph attributes<br />
<br />
The four radio buttons on the left-hand side are used to specify what kind of management one is looking for, namely<br />
* configuration management (manually)<br />
* value management (manually)<br />
* operation involving value management (automatically)<br />
* import & export management (automatically)<br />
<br />
<br />
== manual modifications ==<br />
<br />
Note, that attribute values are typically not added manually (which would be tiresome). It is more usual that attributes are [[#import_and_export|imported]] from files or computed by a [[analysis_tab|network analysis algorithm]]...<br />
<br />
=== configuration ===<br />
<br />
* check mark: the check mark in the left column defines the attribute that is displayed as label (for nodes/links)<br />
* name field: enables the (re-)definition of attribute names; to create a new attribute you have to use this field to define a name first<br />
* type selection: defines the type (text, binary, integer, decimal, text list, integer list, decimal list) of the attribute<br />
* default: can be used to define a default value for this attribute that is assigned to all nodes automatically (since only declaring the name and type of an attribute function does not define its values)<br />
* description<br />
<br />
Don't forget to press the apply button at the bottom of the attribute manager.<br />
<br />
<br />
=== values ===<br />
<br />
The values category can be used to explore the values that the nodes/links (or the selected nodes/links, if there are selected nodes/links) assume on all attribute functions that are declared. You can also modify the values if the '''allow editing''' box at the bottom right of the attribute manager is checked. Additionally, you can specify whether to '''show default values''' (bottom left) and whether to display only a subset of attributes (use check marks at the top).<br />
<br />
Values that are ''not assigned'' yet, are marked ''N/A''.<br />
<br />
<br />
== automatical modifications ==<br />
<br />
=== operations ===<br />
<br />
The operations category provides means to change a node/link attribute function as a whole.<br />
<br />
You might just want to '''copy''' or '''delete''' (select according ''operation'') an existing attribute function (select according ''attribute'') - and provide a new name of the ''result attribute'' (if you don't want to delete or, in general, overwrite the current one).<br />
<br />
You can also '''rank''' the nodes/links in '''ascending''' or '''descending''' ''order'' with regard to an ''attribute'', creating an integer type ''result attribute''.<br />
<br />
'''Merge to list''' enables you to select a number of ''source attributes'' (pressing Crtl key to add single ones or holding shift key to select a number of successive ones) to be merged into a text ''result attribute'', being a comma separated list of these values.<br />
<br />
There are also a number of possibilities to '''convert list values''', namely calculating the '''maximum''', '''minimum''', '''average''', or '''sum''' of entries (integer/decimal list), or '''concatenate''' them (text list).<br />
<br />
Finally, '''manipulate values''' provides means to automatically<br />
* '''invert''' (note, that a zero attribute value will result in a N/A entry in the result attribute!)<br />
* '''reverse''' <br />
* '''add''' a specified ''offset''<br />
* '''scale''' with a specified ''scalar''<br />
* '''normalize''' (i.e. each value is divided by the maximum values)<br />
* '''standardize''' (i.e. each value is divided by the sum of all values)<br />
* '''round''' to specified ''fraction digits''<br />
* '''round up''' to specified ''fraction digits''<br />
the values of an attribute function.<br />
<br />
=== import and export ===<br />
<br />
When importing attributes from a .csv file you have to specify how the data shall be joined.<br />
That is, the .csv file should contain a column that is named according to the '''join by''' attribute of the network.<br />
It is also possible to import an additional link attribute via a .csv matrix (nodes IDs in the matrix have to be the same as in the network).<br />
<br />
When exporting attributes you can specify how the data shall be sorted, using the attribute selected for '''sort by'''.<br />
<br />
If you are working with more than one network tabs, you can also export one specified attribute of all open networks as a table.</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=Main_Page&diff=247Main Page2011-01-21T13:09:38Z<p>Mader: </p>
<hr />
<div>visone (''italian for mink'')<br />
is a software for the '''visual''' creation, transformation,<br />
exploration, analysis, and representation of network data,<br />
jointly developed at the University of Konstanz and the Karlsruhe Institute of Technology since 2001.<br />
<br />
<br />
To provide easy and fast access for newbies and support advanced users equally, the visone wiki <br />
is built on these main categories:<br />
<br />
* [[Trails]] guide you through '''common usage scenarios'''; they are probably the easiest way to learn about visone <br />
<br />
* Complementarily, a '''quick reference guide''' is describing<br />
** [[GUI|Graphical User Interface]] components and their functionalities and <br />
** [[Concepts|Algorithms, Methods, Concepts]] underlying the implementation<br />
<br />
* [[Applications|Data sets and research applications]] provide additional guidance future applications can build on<br />
<br />
<br />
{| id="welcome" style="width:100%; background:#f9f9f9; margin:1.2em 0 6px 0; border:1px solid #ddd;"<br />
| style="white-space:nowrap; color:#000; font-size:162%; border:none; margin:0; padding:1em;" | Welcome to visoneWiki<br />
|-<br />
| style="padding:2px 5px;" |<br />
visone (''italian for mink'') is a software for the '''visual''' creation, transformation, exploration, analysis, and representation of network data, jointly developed at the University of Konstanz and the Karlsruhe Institute of Technology since 2001.<br />
<br />
To provide easy and fast access for newbies and support advanced users equally, the visone wiki is built on these main categories:<br />
|}<br />
<br />
<br />
{| id="overview" style="width: 100%; margin:4px 0 0 0; background:none; border-spacing: 0px;"<br />
<br />
| style="border:1px solid #cef2e0; background:#f5fffa; vertical-align:top; color:#000;" |<br />
{| id="trails" style="width:100%; vertical-align:top; background:#f5fffa;"<br />
! style="padding:2px;" | <h2 style="margin:3px; background:#cef2e0; font-size:120%; font-weight:bold; border:1px solid #a3bfb1; text-align:left; color:#000; padding:0.2em 0.4em;">Trails</h2><br />
|-<br />
| style="color:#000; padding:2px 5px" | <br />
<div><br />
[[Trails]] guide you through '''common usage scenarios'''; they are probably the easiest way to learn about visone<br />
</div><br />
|}<br />
<br />
|-<br />
| style="border:1px solid transparent;" |<br />
|-<br />
<br />
| style="border:1px solid #cedff2; background:#f5faff; vertical-align:top;"|<br />
{| id="quickreference" style="width:100%; vertical-align:top; background:#f5faff;"<br />
! style="padding:2px;" | <h2 style="margin:3px; background:#cedff2; font-size:120%; font-weight:bold; border:1px solid #a3b0bf; text-align:left; color:#000; padding:0.2em 0.4em;">Quick reference guide</h2><br />
|-<br />
| style="color:#000; padding:2px 5px;" | <br />
<div><br />
Complementarily, a '''quick reference guide''' is describing<br />
* [[GUI|Graphical User Interface]] components and their functionalities and <br />
* [[Concepts|Algorithms, Methods, Concepts]] underlying the implementation<br />
</div><br />
|}<br />
<br />
|-<br />
| style="border:1px solid transparent;" |<br />
|-<br />
<br />
| style="border:1px solid #ddcef2; background:#faf5ff; vertical-align:top; color:#000;"|<br />
{| id="application" style="width:100%; vertical-align:top; background:#faf5ff; color:#000; width:100%"<br />
! style="padding:2px;" | <h2 style="margin:3px; background:#ddcef2; font-size:120%; font-weight:bold; border:1px solid #afa3bf; text-align:left; color:#000; padding:0.2em 0.4em">Data sets and research applications</h2><br />
|-<br />
| style="color:#000; padding:2px 5px 5px;" | <br />
<div><br />
[[Applications|Data sets and research applications]] provide additional guidance future applications can build on<br />
</div><br />
|}<br />
<br />
|}</div>Maderhttps://visone.ethz.ch/wiki/index.php?title=File:Algo_logo.png&diff=237File:Algo logo.png2011-01-14T16:53:02Z<p>Mader: algo logo</p>
<hr />
<div>algo logo</div>Mader