The effect of different context representations on word sense discrimination in biomedical texts

Unsupervised word sense discrimination relies on the idea that words that occur in similar contexts will have similar meanings. These techniques cluster multiple contexts in which an ambiguous word occurs, and the number of clusters discovered indicates the number of senses in which the ambiguous word is used. One important distinction among these methods is the underlying means of representing the contexts to be clustered. This paper compares the efficacy of first-order methods that directly represent the features that occur in a context with several second-order methods that use a more indirect representation. The experiments in this paper show that second order methods that use word by word co-occurrence matrices result in the highest accuracy and most robust word sense discrimination. These experiments were conducted on MedLine abstracts that contained pseudo--words created by conflating together pairs of MeSH preferred terms to create new ambiguous words. The experiments were carried out with SenseClusters, a freely available open source software package.