Automatic thesaurus construction using Bayesian networks
Young C. Park, Young Soo Han, Key‐Sun Choi · 1995
Automaticthesaurus construction is accomplished by extracting term relations mechanically.A popular method uses statistical anaJysis to discover the term relations.For low frequency terms, however, the statistical information of the terms cannot be reliably used for deciding the relationship of terms.This problem is generally referred as data sparseness problem.Unfortunately, many studies have been said that the low frequency terms are of most use in information retrieval.In this paper, we formalize the statistical behaviors of terms by using inference network.A formal approach for the data sparseness problem which is crucial in constructing a thesaurus is developed.The validity of this approach is shown by experiments.