An interactive technique for the analysis of multivariate data

Abstract

IDEA, a computer program for Inductive Data Exploration and Analysis, is operational under SDC's 0-32 time-sharing system. Using IDEA, a researcher, through on-line interaction with his data, is able to explore various ways of accounting for relationships that exist among his measurements. Instead of conventional poly-nominal regression coefficients, IDEA represents these relationships as decision trees whose form, shape, depth and so on are determined by the user from his data, using unique IDEA heuristics to guide him. Three case studies are reported in which IDEA is contrasted with discriminant analysis. A summary is made of a preliminary study using artificial data of the effect of measurement error on IDEA analysis. OCT)DEA, a system of computer programs for 364

Laboratory experimentation at the center for computer-based behavior studies

“…These graphically represent the rule used to partition a set of observations into exhaustive, mutually exclusive regions. IDEA was developed for on-line interaction in a time-shared computer environment through hard copy terminals or cathode ray tube displays to permit the investigator to collaborate with a library of programmed heuristics in the process of uncovering and representing the structure in his data (Shure, Press, & Rogers, 1968;Press, Rogers, & Shure, 1969).…”

The use of time-shared terminals in psychology

“…Compu ter systems have been devised for more elaborate interactive data analysis (Press, Rogers, & Shure, 1969). Generally, these systems are not available on commercial time-shared facilities, since they require very large computers to run.…”

Exploratory Analysis of Marketing Data: Trees vs. Regression

“…Support for the benefits of trees over regression has been provided with simulated data [10] and with actual data [2,8,9,10]. However, the studies which used actual data have been restricted to a comparison of the ability to fit the model to the data.…”