论文信息 - SenseClusters - Finding Clusters that Represent Word Senses

SenseClusters - Finding Clusters that Represent Word Senses

SenseClusters is a freely available word sense discrimination system that takes a purely unsu-pervised clustering approach. It uses no knowledge other than what is available in a raw unstructured corpus, and clusters instances of a given target word based only on their mutual contextual similarities. It is a complete system that provides support for feature selection from large corpora, several different context representation schemes, various clustering algorithms, and evaluation of the discovered clusters.

Ted Pedersen | Amruta Purandare

[1] Amruta Purandare. Discriminating Among Word Senses Using McQuitty's Similarity Analysis , 2003, HLT-NAACL.

[2] Ted Pedersen,et al. Word Sense Discrimination by Clustering Contexts in Vector and Similarity Spaces , 2004, CoNLL.

[3] Richard A. Harshman,et al. Indexing by Latent Semantic Analysis , 1990, J. Am. Soc. Inf. Sci..

[4] George Karypis,et al. CLUTO - A Clustering Toolkit , 2002 .

[5] Satanjeev Banerjee,et al. The Design, Implementation, and Use of the Ngram Statistics Package , 2003, CICLing.

[6] Hinrich Schütze,et al. Automatic Word Sense Discrimination , 1998, Comput. Linguistics.

[7] G. Miller,et al. Contextual correlates of semantic similarity , 1991 .

[8] Ted Pedersen,et al. Distinguishing Word Senses in Untagged Text , 1997, EMNLP.