17.txt

This complete matlab for neural network

发信人: GzLi (笑梨), 信区: DataMining
标  题: two ML software zz
发信站: 南京大学小百合站 (Wed Jun  4 08:40:55 2003)

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

-SLIPPER: http://software.cs.rutgers.edu/slipper/slipper_license.html
-WHIRL: http://software.cs.rutgers.edu/whirl/whirl_license.html

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
-- 
--------------
William W. Cohen
[e-mail]
http://www.wcohen.com
Senior Research Scientist
CALD, Carnegie Mellon University
--
              ***  端庄厚重 谦卑含容 事有归着 心存济物  ***
数据挖掘  http://[e-mail]/

※ 来源:．南京大学小百合站 bbs.nju.edu.cn．[FROM: 202.120.8.48]