Paper: Distributional Measures Of Concept-Distance: A Task-Oriented Evaluation

ACL ID W06-1605
Title Distributional Measures Of Concept-Distance: A Task-Oriented Evaluation
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2006
Authors

We propose a framework to derive the distance between concepts from distribu- tional measures of word co-occurrences. We use the categories in a published thesaurus as coarse-grained concepts, al- lowing all possible distance values to be stored in a concept–concept matrix roughly.01% the size of that created by existing measures. We show that the newly proposed concept-distance mea- sures outperform traditional distributional word-distance measures in the tasks of (1) ranking word pairs in order of se- mantic distance, and (2) correcting real- word spelling errors. In the latter task, of all the WordNet-based measures, only that proposed by Jiang and Conrath out- performs the best distributional concept- distance measures. 1 Semantic and distributional measures Measures of distance of m...