The 3DVDM is a system for the analysis of multi-dimensional data. In their latest version in 2008, forest cover data obtained from the US Forest Services where analysed through a CAVE. The system allows data to be represented through a 3D Scatterplot or a 3D SPLOM (Scatterplot of Matrices). Moreover, color, size and shape are used to encode additional attributes.
The particularity of the 3DVDM approach however, is in the use of additional channel such as dynamic object property, hence blinking, rotating, or vibrating a point. On top of that, sound was used to encode categorical or numerical attributes. Sound attributes are only triggered within a certain range of the user. Categorical attributes are displayed through spoken number and numerical one through a pitch interpolation.
Sources
[1] H. R. Nagel, E. Granum, and P. Musaeus, “Methods for Visual Mining of Data in Virtual Reality,” Proceedings of the International Workshop on Visual Data Mining, pp. 13–27, 2001.
[2] H. R. Nagel, M. Vittrup, E. Granum, and S. Bovbjerg, “Exploring Non-Linear Data Relationships in VR using the 3D visual data mining system,” in Third International Workshop on Visual Data Mining in Conjunction With Icdm 2003-the Third Ieee International Conference on Data Mining, Melbourne, Fl, US, 2003, pp. 133–150.
[3] H. R. Nagel, E. Granum, S. Bovbjerg, and M. Vittrup, “Immersive Visual Data Mining: The 3DVDM Approach,” in Visual Data Mining. Berlin, Heidelberg: Springer Berlin Heidelberg, 2008, pp. 281–311.