A Simple Rule-Based Part of Speech Tagger

A SIMPLE RULE-BASED PART OF SPEECH TAGGER Eric Brill * Department of Computer Science University of Pennsylvania Philadelphia, Pennsylvania 19104 brill~unagi.cis.upenn.edu ABSTRACT Automatic part of speech tagging is an area of natural lan- guage processing where statistical techniques have been more successful than rule-based methods. In this paper, we present a simple rule-based part of speech tagger which automati- cally acquires its rules and tags with accuracy comparable to stochastic taggers. The rule-based tagger has many ad- vantages over these taggers, including: a vast reduction in stored information required, the perspicuity of a small set of meaningful rules, ease of finding and implementing im- provements to the tagger, and better portability from one tag set, corpus genre or language to another. Perhaps the biggest contribution of this work is in demonstrating that the stochastic method is not the only viable method for part of speech tagging. The fact that a simple rule-based tagger that automatically learns its rules can perform so well should offer encouragement for researchers to further explore rule- based tagging, searching for a better and more expressive set of rule templates and other variations on the simple but effective theme described below. 1. INTRODUCTION There has been a dramatic increase in the application of probabilistic models to natural language processing over the last few years. The appeal of stochastic techniques over traditional rule-based techniques comes from the ease with which the necessary statistics can be automat- ically acquired and the fact that very little handcrafted knowledge need be built into the system. In contrast, the rules in rule-based systems are usually difficult to construct and are typically not very robust. One area in which the statistical approach has done par- ticularly well is automatic part of speech tagging, as- signing each word in an input sentence its proper part of speech [1, 2, 3, 4, 6, 9, 11, 12]. Stochastic taggers have *A version of this paper appears in Proceedings of the Third Conference on Applied Computational Linguistics (ACL), Trento, Italy, 1992. Used by permission of the Association for Computa- tional Linguistics; copies of the publication from which this ma- terial is derived can can be obtained from Dr. Donald E. Walker (ACL), Bellcore, MRE 2A379, 445 South Street, Box 1910, Morris- town, NJ 07960-1910, USA. The author would like to thank Mitch Marcus and Rich Pito for valuable input. This work was supported by DARPA and AFOSR jointly under grant No. AFOSR-90-0066, and by ARO grant No. DAAL 03-89-G0031 PRI. 112 obtained a high degree of accuracy without performing any syntactic analysis on the input. These stochastic part of speech taggers make use of a Markov model which captures lexical and contextual information. The parameters of the model can be estimated from tagged [1, 3, 4, 6, 12] or untagged [2, 9, 11] text. Once the parameters of the model are estimated, a sentence can then be automatically tagged by assigning it the tag se- quence which is assigned the highest probability by the model. Performance …

Content truncated to 500 words