DATA-COMPRESSION IN MACHINE LEARNING APPLIED TO NATURAL-LANGUAGE

Authors
GRZYMALABUSSE JW THAN S
Citation
Jw. Grzymalabusse et S. Than, DATA-COMPRESSION IN MACHINE LEARNING APPLIED TO NATURAL-LANGUAGE, Behavior research methods, instruments, & computers, 25(2), 1993, pp. 318-321
Citations number
12
Categorie Soggetti
Psychology, Experimental
Journal title
Behavior research methods, instruments, & computers → ACNP
ISSN journal
07433808
Volume
25
Issue
2
Year of publication
1993
Pages
318 - 321
Database
ISI
SICI code
0743-3808(1993)25:2<318:DIMLAT>2.0.ZU;2-H
Abstract
In this paper, we investigate the possibility of applying machine lear ning methods to data derived from the area of natural language and sho w how rules, induced by machine learning, are changed after the origin al data are compressed by grouping together entries, attributes, and a ttribute values. Also shown is how excessive compression of input data may affect the accuracy of induced rules.