This paper describes a program that disambiguates English word senses in unrestricted text using statistical models of the major Roget's Thesaurus categories. Roget's categories serve as approximations of conceptual classes. The categories listed for a word in Roget's index tend to correspond to sense distinctions; thus selecting the most likely category provides a useful level of sense disambiguation. The selection of categories is accomplished by identifying and weighting words that are indicative of each category when seen in context, using a Bayesian theoretical framework. Other

Abstract not found

Most probabilistic classifiers used for word-sense disambiguation have either been based on only one contextual feature or have used a model that is simply assumed to characterize the interdependencies among multiple contextual features. In this paper, a different approach to formulating a probabilistic model is presented along with a case study of the performance of models produced in this manner for the disambiguafion of the noun interest. We describe a method for formulating probabilistic models that use multiple contextual features for word-sense disambiguafion, without requiring untested assumptions regarding the form of the model. Using this approach, the joint distribution of all variables is described by only the most systematic variable interactions, thereby limiting the number of parameters to be estimated, supporting computational efficiency, and providing an understanding of the data.

The automatic disambiguation of word senses has been an interest and concern since the earliest days of computer treatment of language in the 1950's. Sense disambiguation is an “intermediate task ” (Wilks and Stevenson, 1996) which is not an end in itself, but rather is necessary at one level or another to accomplish most natural language processing tasks. It is

Word sense disambiguation has been recognized as a major problem in natural language processing research for over forty years. Much of this work has been stymied by difficulties in acquiring appropriate lexical resources, such as semantic networks and annotated corpora. Following the suggestion in Brown et al. (1991a) and Dagan et al. (1991), we have achieved considerable progress recently by taking advantage of a new source of testing and training materials. Rather than depending on small amounts of hand-labeled text, we have been making use of relatively large amounts of parallel text, text such as the Canadian Hansards (parliamentary debates), which are available in two (or more) languages. The translation can often be used in lieu of hand-labeling. For example, consider the polysemous word sentence, which has two major senses: (1) a judicial sentence, and (2), a syntactic sentence. We can collect a number of sense (1) examples by extracting instances that are translated as peine, and we can collect a number of sense (2) examples by extracting instances that are translated as phrase. In this way, we have been able to acquire a considerable amount of testing and training material for developing and testing our disambiguation algorithms. The availability of this testing and training material has enabled us to develop quantitative disambiguation methods that achieve 90 % accuracy in discriminating between two very distinct senses of a noun such as

Discrimination decisions arise in many natural language processing tasks. Three classical tasks are discriminating texts by their authors (author identification), discriminating documents by their relevance to some query (information retrieval), and discriminating multi-meaning words by their meanings (sense discrimination). Many other discrimination tasks arise regularly, such as determining whether a particular proper noun represents a person or a place, or whether a given word from some teletype text would be capitalized if both cases had been used. We (1992) introduced a method designed for the sense discrimination problem. Here we show that this same method is useful in each of the five text discrimination problems mentioned. We also discuss areas for research based on observed shortcomings of the method. In particular, an example in the author identification task shows the need for a robust version of the method. Also, the method makes an assumption of independence which is demon...