CViz

A visualization tool designed for analyzing high-dimensional data in large, complex data sets.

Date Posted: June 15, 1998

Overview

Requirements

Update: November 19, 2003

New release extends the expiration date to 12/31/2005.

CViz is rated

What is CViz?

CViz is a visualization tool designed for analyzing high-dimensional data (data with many elements) in large, complex data sets. CViz easily loads the data sets, displays the most important factors relating clusters of records, and provides full-motion visualization of the inherent data clusters.

How does it work?

This tool is very useful for analysts who often use statistical methods. It helps them decide on the best factors to use in multiple regression analysis and other statistical methods, primarily because it is based on linear discriminant analysis and it conforms to rigorous standards in statistical mathematics.

About the technology author(s):
David C. Martin
Scott Spangler
Dharmendra S. Modha
Shivakumar Vaithyanathan
Inderjit S. Dhillon

David C. Martin is a senior development manager with IBM's Global Business Intelligence Solutions (GBIS) unit. Mr. Martin has been working in data and knowledge management for over 15 years and currently leads GBIS's advanced technology group based at the IBM Almaden Research Center in San Jose, California. Mr. Martin previously held positions at the University of California, Sun Microsystems, and a number of Silicon Valley start-ups. Mr. Martin is a member of IEEE, ACM, and the Internet Society.

Scott Spangler Scott Spangler holds a Bachelor's degree in Math from MIT and a Master's degree in Computer Science from the University of Texas (Austin). He is currently a research member with the net.Mining R&D group of the Global Business Intelligence Solutions department, located at the IBM Almaden Research Center. Before that, Mr. Spangler worked at the General Motors Tech Center for ten years on data mining and knowledge-based systems development. In addition to CViz, he developed systems for predicting Web surfer behavior as well as statistical design of experiments. He won GM's "Boss" Kettering Award for technical innovation, and he has one patent from GM and has applied for two patents at IBM. Since designing CViz, Mr. Spangler has recently become involved in several Data Visualization efforts within GBIS and IBM Research.

Dharmendra S. Modha obtained his Ph.D. in Electrical and Computer Engineering from University of California at San Diego in 1995. He is currently a research member with the net.Mining R&D group of the Global Business Intelligence Solutions department, located at the IBM Almaden Research Center. His work at IBM has resulted in four patent applications in clustering, dimensionality reduction, visualization, and data mining on parallel machines. He has more than 15 journal and conference papers to his credit. Dr. Modha is a member of the Institute of Electrical and Electronics Engineers and the Institute for Mathematical Statistics.

Shivakumar Vaithyanathan is currently a research member with the net.Mining R&D group of the Global Business Intelligence Solutions department, located at the IBM Almaden Research Center. After obtaining his Ph.D. in 1992, Dr. Vaithyanathan was a visiting scientist at Lehigh University before he joined Digital Equipment Corp. to work on advanced algorithms for process control. Subsequently, he moved to the newly-formed AltaVista group and was responsible for the development of document-clustering algorithms. Since joining IBM in 1996, he has been involved in research and development of dimensionality reduction algorithms and unsupervised learning algrithms, especially for extremely high-dimensional sparse data. Dr. Vaithyanathan's present interests are in the area of bayesian inference and maximum entropy models, user modeling and profiling, and partially-supervised learning algorithms.

Inderjit S. Dhillon obtained his Ph.D. in Computer Sience from the University of California at Berkeley in 1997. He is currently a research member with the net.Mining R&D group of the Global Business Intelligence Solutions department, located at the IBM Almaden Research Center. Dr. Dhillon's thesis work has culminated in the world's fastest symmetric eigensolver and will soon be available through the LAPACK library. In his current data-mining work at IBM, Dr. Dhillon is mainly interested in efficient and scalable solutions to dimensionality reduction and in automatic categorization and visualization of extremely high-dimensional sparse data. Dr. Dhillon has contributed more than 15 journal and conference papers and has given several talks at international conferences. He is a SIAM (Society of Industrial and Applied Mathematics) member.