Visualization and Visual Analysis Workbench is a Java™-based workbench that provides a large selection of data visualizations and a comprehensive library of mathematical and statistical functions. The user can create and manipulate multiple interactive visualizations and can create new variables and transform variables "on the fly." With this workbench, the user can easily gain insight into data by using dynamically linked visualizations with color marking and queries.

Visualization and Visual Analysis Workbench supports interactive visualization and analysis, including dynamic linking between visualizations. The use of color to mark regions in one visualization automatically causes corresponding data regions in all the other visualizations to be highlighted. This application works interactively on large data sets with more than a million rows of data.

How does it work?

With Visualization and Visual Analysis Workbench, data sets can be loaded from files or from the clipboard. The data files are in simple, delimited row and column format and can be cut and pasted from a spreadsheet. The user dynamically selects visualizations, defines variables, and selects transforms. To get started, the user interactively explores the data by launching any number of visualizations.

In all visualizations, the user can mark regions of interest with color. Those cases are automatically colored the same way in all other visualizations and data tables that the user is examining or that he later launches. The user can view each visualization in different ways, such as by adding voronoi diagrams, error bars, or scatterplots. Various coloring mechanisms are provided.

The workbench supports visualization of data values and of images associated with data values. Example visualizations include scatterplots, pie charts, histograms, parallel coordinates, tree maps, dendograms (correlation trees) for variables and cases, automatic computation of univariate and bivariate statistics, and so on.

David Rabenhorst created Visualization and Visual Analysis Tool and its C++ predecessor. These software libraries and end-user applications provided ground-breaking interactive visual analysis and visualization of tabular data, including new visualization and analysis metaphors. Mr. Rabenhorst created this tool as a member of the Visual Analysis Group at IBM Research, where he filed patents on the interactive analysis of many-dimensional data. Mr. Rabenhorst is an electrical engineer and is now retired from IBM.

Bernice Rogowitz was the manager of the Visual Analysis Group. She contributed visual analysis methods based on human perception. Dr. Rogowitz is an experimental psychologist working in the Collaborative User Experience group at IBM Research.

Genady Grabarnik, Ph.D., is a mathematician at IBM Research; he provided support and extensions for this tool.