A New Approach to Word Sense Disambiguation

A New Approach to Word Sense Disambiguation Rebecca Bruce and Janyce Wiebe The Computing Research Lab New Mexico State University Las Cruces, NM 88003 ABSTRACT This paper presents and evaluates models created according to a schema that provides a description of the joint distribu- tion of the values of sense tags and contextual features that is potentially applicable to a wide range of content words. The models are evaluated through a series of experiments, the results of which suggest that the schema is particularly well suited to nouns but that it is also applicable to words in other syntactic categories. 1. INTRODUCTION Assigning sense tags to the words in a text can be viewed as a classification problem. A probabilistic classifier assigns to each word the tag that has the highest estimated proba- bility of having occurred in the given context. Designing a probabilistic classifier for word-sense disambiguation includes two main sub-tasks: specifying an appropriate model and estimating the parameters of that model. The former in- volves selecting informative contextual features (such as col- locations) and describing the joint distribution of the values of these features and the sense tags of the word to be classi- fied. The parameters of a model are the characteristics of the entire population that are cohsidered in the model. Practical applications require the use of estimates of the parameters. Such estimates are based on functions of a data sample (i.e., statistics) rather than the complete population. To make the estimation of parameters feasible, a model with a simplified form is created by limiting the number of contextual features considered and by expressing the joint distribution of fea- tures and sense tags in terms of only the most important systematic interactions among variables. To date, much of the work in statistical NLP has focused on parameter estimation ([11], [13], [12], [4]). Of the research di- rected toward identifying the optimum form of model, most has been concerned with the selection of individually infor- mative features ([2], [5]), with relatively little attention di- rected toward the identification of an optimum approxima- tion to the joint distribution of the values of the contextual features and object classes. Most previous efforts to formu- late a probabilistic classifier for word-sense disambiguation did not attempt to systematically identify the interdepen- dencies among contextual features that can be used to clas- sify the meaning of an ambiguous word. Many researchers have performed disambiguation on the basis of only a single feature ([61, [15], [2]), while others who do consider multiple contextual features assume that all contextual features are either conditionally independent given the sense of the word (Is], [14]) or fuRRy independent ([10], [16]). J In earlier work, we describe a method for identifying an up- propriate model for use in disambiguating a word given a set of contextual features. We chose a particular set of con- textual features and, using this method, identified a model incorporating these features for use in disambiguating the noun interest. These features, which are assigned …

Content truncated to 500 words