Extending the dimensionality of flatland with attribute view probabilistic models

In much of Bertin's Semiology of Graphics, marks representing individuals are arranged on paper according to their various attributes (components). Paper and computer monitors can conveniently map two attributes to width and height, and can map other attributes into nonspatial dimensions such as texture, or colour. Good visualizations exploit the human perceptual apparatus so that key relationships are quickly detected as interesting patterns. Graphical models take a somewhat dual approach with respect to the original information. Components, rather than individuals, are represented as marks. Links between marks represent conceptually simple, easily computable, and typically probabilistic relationships of possibly varying strength, and the viewer studies the diagram to discover deeper relationships. Although visually annotated graphical models have been around for almost a century, they have not been widely used. We argue that they have the potential to represent multivariate data as generically as pie charts represent univariate data. The present work suggests a semiology for graphical models, and discusses the consequences for information visualization.

[1]  P. Spirtes,et al.  Causation, prediction, and search , 1993 .

[2]  Ben Shneiderman,et al.  Direct Manipulation: A Step Beyond Programming Languages , 1983, Computer.

[3]  B. Marx.  The Visual Display of Quantitative Information , 1985 .

[4]  Alfred Inselberg,et al.  The automated multidimensional detective , 1999, Proceedings 1999 IEEE Symposium on Information Visualization (InfoVis'99).

[5]  N. Cartwright.  In Favor of Laws that Are Not Ceteris Paribus After All , 2002 .

[6]  Jiji Zhang,et al.  Adjacency-Faithfulness and Conservative Causal Inference , 2006, UAI.

[7]  P. Fayers,et al.  The Visual Display of Quantitative Information , 1990 .

[8]  Robert Spence,et al.  Externalising abstract mathematical models , 1996, CHI '96.

[9]  Colin Ware,et al.  Information Visualization: Perception for Design , 2000 .

[10]  Manon J. Sanscartier,et al.  Exploring causal influences , 2005, IS&T/SPIE Electronic Imaging.

[11]  L. Dennis,et al.  Sunscreen Use and the Risk for Melanoma: A Quantitative Review , 2003, Annals of Internal Medicine.

[12]  Peter L. Brooks,et al.  Visualizing data , 1997 .

[13]  E. Tufte.  Beautiful Evidence , 2006 .

[14]  S. Wright,et al.  The Relative Importance of Heredity and Environment in Determining the Piebald Pattern of Guinea-Pigs. , 1920, Proceedings of the National Academy of Sciences of the United States of America.

[15]  Steffen L. Lauritzen,et al.  Stable local computation with conditional Gaussian distributions , 2001, Stat. Comput..

[16]  Edward R. Tufte,et al.  Envisioning Information , 1990 .

[17]  John J. Bertin,et al.  The semiology of graphics , 1983 .

[18]  S. Geneletti.  Identifying direct and indirect effects in a non‐counterfactual framework , 2007 .

[19]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[20]  Marek J. Druzdzel,et al.  What Do College Ranking Data Tell Us About Student Retention , 1994 .

[21]  K. Gabriel,et al.  The biplot graphic display of matrices with application to principal component analysis , 1971 .