Polysemy and Category Structure in WordNet: An Evidential Approach

While polysemy is a form of ambiguity that can complicate natural language processing, it is also a rich lexical resource that yields useful insights into the mapping between words and concepts. WordNet, a comprehensive lexical knowledge-base of English word meanings, is replete with instances of polysemy, but also contains many instances of homonymy, and fails to distinguish between both kinds of ambiguity. We propose in this paper an alternative to the distributional approach for recognizing polysemous sense-pairs in WordNet. Our approach does not rely on the systematicity of regular polysemy to identify the families of words that instantiate a particular metonymic pattern, but seeks instead local ontological evidence for each word, on a case by case basis.