Paper: Word Sense Disambiguation using Optimised Combinations of Knowledge Sources

ACL ID P98-2228
Title Word Sense Disambiguation using Optimised Combinations of Knowledge Sources
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 1998
Authors

Word sense disambiguation algorithms, with few ex- ceptions, have made use of only one lexical know- ledge source. We describe a system which performs word sense disambiguation on all content words in free text by combining different knowledge sources: semantic preferences, dictionary definitions and sub- ject/domain codes along with part-of-speech tags, optimised by means of a learning algorithm. We also describe the creation of a new sense tagged corpus by combining existing resources. Tested accuracy of our approach on this corpus exceeds 92%, demonstrat- ing the viability of all-word disambiguation rather than restricting oneself to a small sample.