Khmer WordNet construction

Abstract

This thesis described a semi-automatic translation-based expansion approach for Khmer WordNet construction. This approach expanded the Princeton English WordNet's synsets and took into account the existing machine-readable Cambodian-English dictionary. The approach started with the cleansing and extracting the translation links and the semantic links from the Cambodian-English dictionary and the Princeton WordNet, respectively. Concerning the translation links and the semantic links, a large number of candidate links between Khmer words and English synsets can be derived. This approach applied a statistical approach sampling these candidate links and constructed a statistical model based on human verification of the sample of the candidate links. Finally, the statistical model is applied to construct the Khmer WordNet with promising coverage and accuracy. This Khmer WordNet covered 3,402 Khmer words, 3,571 synsets, 6,137 links, and 13.86% core concept coverage. The Khmer WordNet Web-based application has been developed.