Abstract

Two methods for learning text classifiers are compared on classification problems that might arise in filtering and filing personal e-mail messages: a "traditional IR" method based on TF-IDF weighting, and a new method for learning sets of "keyword-spotting rules" based on the RIPPER rule learning algorithm. It is demonstrated that both methods obtain significant generalizations from a small number of examples; that both methods are comparable in generalization performance on problems of this type; and that both methods are reasonably efficient, even with fairly large training sets. However, the greater comprehensibility of the rules may be advantageous in a system that allows users to extend or otherwise modify a learned classifier. Introduction Perhaps the most-discussed technical phenomenon of recent years has been the rapid growth of the Internet---or more generally, the rapid growth in the number of on-line documents. This has led to increased interest in intelligent methods for ...

Citations

800	Fast effective rule induction - Cohen - 1995
784	Learning logical definitions from relations - Quinlan - 1989
290	WebWatcher: A Learning Apprentice for the World Wide Web. AAAI Spring Symposium on Information Gathering from Heterogeneous, Distributed Environments - Armstrong, Freitag, et al. - 1995
239	Comparison of Two Learning Algorithms for Text Categorization - Lewis, Ringuette - 1994
194	Heterogeneous uncertainty sampling for supervised learning - Lewis, Catlett - 1994
141	Representation and Learning in Information Retrieval - Lewis - 1992
53	Text Categorization and Relational Learning - Cohen - 1995
44	Training Text Classifiers by Uncertainty Sampling - Lewis, Gale - 1994
5	Automated learning of decision rules for text categorization - Weiss - 1994
1	Some experiments in text categorization using rules - Cohen - 1996
1	Text categorization of low quality images - Lang - 1995
1	Relevance feedback information retrieval - Kaufmann - 1994