KDnuggets : News : 2003 : n11 : item11 < PREVIOUS | NEXT >

Software

From: William W. Cohen
Date: 03 Jun 2003
Subject: SLIPPER and WHIRL software available to researchers

Two learning systems written back when I was at AT&T Labs are now available for research purposes, via Rutgers University.

SLIPPER is a rule-learning system based on boosting. The code is based on William Cohen's widely-used RIPPER learning system. Like RIPPER, SLIPPER is fast, robust, and easy to use. SLIPPER also supports set-valued features, which makes it useful for text categorization using a "bag of words" representation of text.

WHIRL is a representation system that combines some of the properties of relational databases, and some of the properties of statistical ranked-retrieval systems. WHIRL can also be used as a nearest-neighbor text classifier.

More background information on SLIPPER and WHIRL can be found at http://wcohen.com/slipper/ and http://wcohen.com/whirl/

My thanks to Haym Hirsh for repeatedly flogging the legal teams at AT&T and Rutgers to make this possible.


William W. Cohen
http://www.wcohen.com
Senior Research Scientist
CALD, Carnegie Mellon University

KDnuggets : News : 2003 : n11 : item11 < PREVIOUS | NEXT >

Copyright © 2003 KDnuggets.   Subscribe to KDnuggets News!