Samuel Brody and Mirella Lapata (2008) Good Neighbors Make Good Senses: Exploiting Distributional Similarity for Unsupervised WSD. In Proceedings of the 22nd International Conference on Computational Linguistics, 65--72. Manchester, UK.

We present an automatic method for sense-labeling of text in an unsupervised manner. The method makes use of distributionally similar words to derive an automatically labeled training set, which is then used to train a standard supervised classifier for distinguishing word senses. Experimental results on the Senseval-2 and Senseval-3 datasets show that our approach yields significant improvements over state-of-the-art unsupervised methods, and is competitive with supervised ones, while eliminating the annotation cost.

@InProceedings{brody-lapata:2008:PAPERS,
  author    = {Brody, Samuel  and  Lapata, Mirella},
  title     = {Good Neighbors Make Good Senses: Exploiting Distributional Similarity for Unsupervised {WSD}},
  booktitle = {Proceedings of the 22nd International Conference on Computational Linguistics (Coling 2008)},
  year      = {2008},
  address   = {Manchester, UK},
  pages     = {65--72}
}