Survey of Word Sense Disambiguation Appr oaches

Document Type


Publication Date



Word Sense Disambiguation (WSD) is an important but challenging technique in the area of natural language processing (NLP). Hundreds of WSD algorithms and systems are available, but less work has been done in regard to choosing the optimal WSD algorithms. This paper summarizes the various knowledge sources used for WSD and classifies existing WSD algorithms according to their techniques. The rationale, tasks, performance, knowledge sources used, computational complexity, assumptions, and suitable applications for each class of WSD algorithms are also discussed. This paper will provide users with general knowledge for choosing WSD algorithms for their specific applications or for further adaptation.


This paper was presented at the 18th International FLAIR Conference in Clearwater Beach, Florida in May, 2005. FLAIR is the Florida Artificial Intelligence Research Society. The paper is freely available from AAAI at http://www.aaai.org/Papers/FLAIRS/2005/Flairs05-051.pdf. Copyright © 2005, American Association for Artificial Intelligence. All rights reserved.