Unsupervised Graph-based Word Sense Disambiguation Using Measures of Word Semantic Similarity Export

@inproceedings{citeulike:4629326, abstract = {This paper describes an unsupervised graph-based method for word sense disambiguation, and presents comparative evaluations using several measures of word semantic similarity and several algorithms for graph centrality. The results indicate that the right combination of similarity metrics and graph centrality algorithms can lead to a performance competing with the state-of-the-art in unsupervised word sense disambiguation, as measured on standard data sets.}, author = {Sinha, Ravi and Mihalcea, Rada}, booktitle = {IEEE International Conference on Semantic Computing (ICSC 2007)}, citeulike-article-id = {4629326}, comment = {Authors propose to disambiguate word senses using the similarities with the surrounding words (with their senses) withing a window. They create a graph with vertexes as senses and weighted edges with similarities. The latter is computed using WordNet and various graph-based and information content similarity measures. The vertexes with the hightest score are used as the most probable senses. There are used several graph-centrality measures for computing score (betweenness, closeness etc.). 67\% precision was reached for nouns in Senseval-2.}, keywords = {disambiguation, semantic\_relatedness, similarity}, posted-at = {2009-05-25 13:01:55}, priority = {0}, title = {Unsupervised Graph-based Word Sense Disambiguation Using Measures of Word Semantic Similarity}, year = {2007} }

TY - CONF ID - citeulike:4629326 L3 - citeulike-article-id:4629326 N2 - This paper describes an unsupervised graph-based method for word sense disambiguation, and presents comparative evaluations using several measures of word semantic similarity and several algorithms for graph centrality. The results indicate that the right combination of similarity metrics and graph centrality algorithms can lead to a performance competing with the state-of-the-art in unsupervised word sense disambiguation, as measured on standard data sets. TI - Unsupervised Graph-based Word Sense Disambiguation Using Measures of Word Semantic Similarity T2 - IEEE International Conference on Semantic Computing (ICSC 2007) KW - disambiguation KW - semantic_relatedness KW - similarity N1 - Authors propose to disambiguate word senses using the similarities with the surrounding words (with their senses) withing a window. They create a graph with vertexes as senses and weighted edges with similarities. The latter is computed using WordNet and various graph-based and information content similarity measures. The vertexes with the hightest score are used as the most probable senses. There are used several graph-centrality measures for computing score (betweenness, closeness etc.). 67% precision was reached for nouns in Senseval-2. AU - Sinha, Ravi AU - Mihalcea, Rada PY - 2007/// ER -