Studies in Part of Speech Labelling

STUDIES IN PART OF SPEECH LABELLING Marie Meteer, Richard Schwartz, Ralph Weischedel BBN Systems and Technologies 10 Moulton St. Cambridge, MA 02138 ABSTRACT We report here on our experiments with POST (Part of Speech Tagger) to address problems of ambiguity and of understanding unknown words. Part of speech tagging, per se, is a well understood problem. Our paper reports experiments in three important areas: handling unknown words, limiting the size of the training set, and returning a set of the most likely tags for each word rather than a single tag. We describe the algorithms that we used and the specific results of our experiments on Wall Street Journal articles and on MUC terrorist messages. 1. INTRODUCTION 1 Natural language processing, and AI in general, have focused mainly on building rule-based systems with carefully handcrafted rules and domain knowledge. Our own natural language database query systems, JANUS 2, Parlance ruz, and Delphi 4, use these techniques quite successfully. However, as we move from the problem of understanding queries in fixed domains to processing open text for applications such as data extraction, we have found rule-based techniques too brittle, and the amount of work necessary to build them intractable, especially when attempting to use the same system on multiple domains. We report in this paper on one application of probabilistic models to language processing, the assignment of part of speech to words in open text. The effectiveness of such models is well known (Church 1988) and they are currently in use in parsers (e.g. de Marcken 1990). Our work is an incremental improvement on these models in two ways: (1) We have run experiments regarding the amount of training data needed in moving to a new domain; (2) we have added probabilistic models of word features to handle unknown words effectively. We describe POST and its algorithms and then we describe our extensions, showing the results of our experiments. 1 The work reported here was supported by the Advanced Research Projects Agency and was monitored by the Rome Air Development Center under Contract No. F30602-87-D-0093. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, whether expressed or implied, of the Defense Advanced Research Projects Agency or the United States Government. 2 Weischedel, et al. 1989. 3 Parlance is a trademark of BBN Systems and Technologies. 4 Stallard, 1989. 2. POST: USING PROBABILITIES TO TAG PART OF SPEECH Predicting the part of speech of a word is one straightforward way to use probabilities. Many words are several ways ambiguous, such as the following: a round table: adjective a round of cheese: noun to round out your interests: verb to work the year round: adverb Even in context, part of speech can be ambiguous, as in the famous example: "Time flies." where both words are two ways ambiguous, resulting in two grammatical interpretations as sentences. Models predicting part of speech can serve to cut down the search space …

Content truncated to 500 words