The SenseRelate project seeks to use measures of semantic similarity and relatedness to perform word sense disambiguation.

We have two different word sense disambiguation algorithms, an "all words" version (WordNet-SenseRelate-AllWords) and a lexical sample version (WordNet-SenseRelate-TargetWord). The all words algorithm assigns a sense to each word in a text, and the lexical sample version assigns a sense to a given target word. There is also a third program which selects the sense of a word that is most related to a given set of words (WordNet-SenseRelate-WordToSet).

All of these assign meanings as found in the lexical database WordNet. They carry out word sense disambiguation by measuring the semantic similarity between a word and its neighbors. In particular, a word is assigned the sense that is most related to its neighbors. The measures of similarity are from wn-similarity, which is also distributed via CPAN as WordNet::Similarity. We have mailing lists for users and news and developers. The mailing lists may be used to discuss any of these packages.


SenseRelate Development Team


The development of SenseRelate has been supported by a National Science Foundation Faculty Early Career Development (CAREER) Program award (#0092784, 2001-2007), by a Grant in Aid of Research, Artistry and Scholarship from the Graduate School of the University of Minnesota (2003-2004), and by the Digital Technology Initiative of the Digital Technology Center of the University of Minnesota (2004-2005).

SourceForge.net Logo CPAN Logo NSF Logo

By: Ted Pedersen - tpederse AT d umn edu