Word sense disambiguation of adjectives using probabilistic networks

In this paper, word sense disambiguation (WSD) ac-curacy achievable by a probabilistic classier, using very minimal training sets, is investigated. We made the assumption that there are no tagged corpora available and identied what information, needed by an accurate WSD system, can and cannot be auto-matically obtained. The lesson learned can then be used to focus on what knowledge needs manual an-notation. Our system, named Bayesian Hierarchical Disambiguator (BHD), uses the Internet, arguably the largest corpus in existence, to address the sparse data problem, and uses WordNet's hierarchy for se-mantic contextual features. In addition, Bayesian networks are automatically constructed to represent knowledge learned from training sets by modeling the selectional preference of adjectives. These net-works are then applied to disambiguation by per-forming inferences on unseen adjective-noun pairs. We demonstrate that this system is able to disam-biguate adjectives in unrestricted text at good initial accuracy rates without the need for tagged corpora. The learning and extensibility aspects of the model are also discussed, showing how tagged corpora and additional context can be incorporated easily to im-prove accuracy, and how this technique can be used to disambiguate other types of word pairs, such as verb-noun and adverb-verb pairs