Corpus-Derived First, Second and Third-Order Word Affinities
Gregory Grefenstette · 1994
A number of corpus-based extraction techniques have been successfully implemented which derive lists of similar words, based on some definition of the context in which they are found, from a corpus. We present here the results of affining such a list in order to extract semantic axes expressing nuances of a word's meaning. These semantic axes represent corpus-based meaning distinctions that are based on the word's usage in the corpus. 1.