Data mining in practice, with STATISTICA
by
R A Hoare
HRS Ltd

Data mining these days consists of two kinds of activity - devising ways to find useful relationships or information buried in very large amounts of data, and devising ways to make those techniques easily accessible to those whose job it is to find the relationships. The developers at StatSoft have concentrated on the latter task, and this paper describes the way in which StatSoft Inc have implemented a wide variety of techniques within STATISTICA Data Miner.

StatSoft have created a graphical data mining interface on top of the standard STATISTICA statistical environment, supplemented by a set of standard and unique data mining tools. All of the analysis features expected in a top-line statistical program are available during data mining, as well as neural networks, various tree-based algorithms, and other procedures which will be announced just prior to the conference. The output tables from data mining are the same as the tables used everywhere else in the program, so can be incorporated in a chain of data analysis without change, saving the user significant time and effort.

Date received: May 10, 2002

Copyright © 2002 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cajj-08.