Automatic Thesaurus

We address the probhml of automaticMly constructing a thesaurus by clustering words based on corpus data. We view this problem as that of estimating a joint distribution over the (:artesian …...