The BeGraph Visualizer tool provides 3D manual exploration and visual configuration capabilities for your data, so you can get an overall description and awareness of your network. In the first place, a node property called Position must be created using one of the layout algorithms. Then the visualizer is active and appears on the screen.
The next window shows an instance of the visualizer with predefined visual configurations for a social network (public Facebook dataset) from our gallery. Here you can try all the functionalities shown below and see how they work.
The visualizer toolbar configures the following options:
|Sets the transparency: opaque nodes and transparent links (default), opaque network and transparent network.|
|Sets the visualizer background to black (moon) or to white (sun). Also changes the way the transparency is treated (aditive VS substractive).|
|Renders or hides self loops (self links) in the network.|
|Toogles node or link selection.|
|Switches ON/OFF the lightning mode, which hides links during network movement. Use it for big networks to speed visual performance.|
|Highlight adjacent nodes mode relative to the current selection: highlights all (default), commons or none of the neighbors.|
|Shadows inactive elements (not selected nor highlighted).|
|Switches on/off network rotation.|
|Shows or hides node/link tags.|
|Isolates selection, hides selection or restores the default full visibility.|
|Visualizer Help, with mouse controls and keyboard shortcuts.|
The visualizer provides a tool to save, expand and recover selections defined by the user. It is located on the lower part of the screen:
Clicking on the numbers lets you either save a selection if the button is empty, or recover a previously saved selection. You can also do it with the keyboard, pressing 'SHIFT + [0..9]' to save a selection and pressing a number to recover.
The two charts on the lower-left corner indicate the number of selected nodes/links (green) and adjacent (purple). Move the mouse over the chart to see the percentages of selected/adjacents.
With a proper visualization, you can roughly identify the most important nodes in the network, the level of clustering, the approximate number of clusters, the strength of the connections between clusters and the global network connectivity.
The visualization is highly configurable. You may set the color, size, and shape of nodes and links according to their properties to emphasize aspects of your visualization and produce aesthetic visualizations to present your results. Those controls are rather intuitive and can be found on the left panel of the application. Check the following pictures to have an idea of the plotting skills:
When the background is black, BeGraph uses an additive transparency (additive blending) model, meaning that transparent elements are considered also light sources. This can produce color saturation, that can be avoided by reducing the element size, their transparency value or increasing the saturation threshold level. When the bakcground color is set to white, the element transparency is treated in a substractive way.
The visualizer is also equipped with an anti-aliasing filter (Fast Approximate Anti-Aliasing) to improve the image quality. If aliasing persists, increasing slightly the size of the nodes/links may reduce it.