Hypernym extraction from Wikipedia
Adel Ghamnia · 2016
Hypernym extraction from Wikipedia The volume of available documents on the Web continues to increase, the texts contained in these documents are rich information describing concepts and relationships between concepts specific to a particular field. In this paper, we propose and exploit an hypernymy extractor based on lexico-syntactic patterns designed for Wikipedia semi-structured pages, especially the disambiguation pages, to enrich a knowledge base as BabelNet and DBPedia. The results show a precision of 0.68 and a recall of 0.75 for the patterns that we have defined, and an enrichment rate up to 33% for both BabelNet and DBPedia semantic resources.