Constructing a Uyghur language semantic lexicon based on WordNet

This paper researches and discusses the English-based WordNet database of semantic relations and proposes a method for constructing a Uyghur language, distributed-system, semantic lexicon. In our work, we first took a modern Uyghur language annotated dictionary containing 60,000 terms, and processed these terms using WordNet´s conceptual relations and thereby extracted from this local language material root words in specific domains. We then used a semi-automated method to create a set of Uyghur synonyms to which Uyghur language information processing could be applied. At present we have compiled a Uyghur language database which includes a total of 1300 pairs of synonyms and 1059 pairs of antonyms.