Child pages
  • 3D Mapping (7.0)



Analysis | Visual | Overall | Mapping | 3D Mapping 


The first Mapping tool was introduced in version 5.1. It has then been updated in version 5.35.4 and 7.0. This tool allows you to select metrics for Nodes and Arcs and have these metrics represented graphically in 2 dimensions as node diameter and/or color, and arc thickness and color respectively.

Mapping in 3 Dimensions 

This new tool is an extension of the 2D Mapping. It thus comes with the same features, but in 3 dimensions. 

The initial layout is the one of the Bayesian network displayed in the Graph pane. It is thus, by default, in 2D. The 3D layout can be generated by clicking the icon ..... or by pressing  as many times as needed to get a satisfying rendering.

Besides the classical tools for zooming, stretching and shrinking the links, two icons are available for manipulating the hidden camera:

  •  to rotate the graph with the mouse while left-clicking
  •  for moving the graph with the mouse while left-clicking

The definition of layers in the 3D layout can be done by associating temporal indices with the nodes.


In the graph below, Purchase Intent has a temporal index of 0, the Factors have a temporal index of 1 and the Manifests a temporal index of 2.


In this example, [Factor_8] has a temporal index of 0, all the other Factors have a temporal index of 1 and the Manifests a temporal index of 2.

Note that the automatic layout 3D layout algorithm does not take into account the defined layers.

Export 3D Model

Whereas the 2D Mapping allows to export the Adjacency Matrix, this mapping allows to generate an XML file describing the 3D model (nodes, arcs, layout and metric values).




As an example, this file is the input of our BayesiaLab Virtual Reality tool. It allows immersing you into your networks and studing them in three dimensions, just like physical objects. Complex networks with hundreds or thousands of nodes, which used to be difficult to comprehend, can now be explored intuitively. All of BayesiaLab's information-theoretic measures, e.g. Mutual Information, Bayes Factor, or Node Force, can be visualized in real-time as you hold and turn your network to view it from all angles.