Word-Sense Disambiguation Using Decomposable Models

Most probabilistic classi ers used for word-sense disambiguationhave either been based on onlyone contextual feature or have used a model that is simply assumed to characterize the interdependencies among multiple contextual features. In this paper, a di erent approach to formulating a probabilistic model is presented along with a case study of the performance of models produced inthis mannerforthe disambiguationofthe noun interest. We describe a method for formulating probabilistic models that use multiplecontextual features for word-sense disambiguation, 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 e ciency, and providing an understanding of the data.

• Type: Conference
• Year: 1994
• Where: ACL
• Authors: Rebecca F. Bruce, Janyce Wiebe
• Keywords: ACL 1994, ACL 2007, Contextual Features, Most Probabilistic Classi, Probabilistic Models