Using WordNet Similarity and Translations to Create Synsets for Ontology-Based Vietnamese WordNet

It is necessary to build a large-scale online lexical-semantic network for the Vietnamese language because it fosters the research and applications in natural language processing, information retrieval, and other areas. The goal of this research is to propose an approach to create Vietnamese WordNet (VWN) automatically. The WordNet similarity was used in order to improve the accuracy of using English WordNet and translations when creating synsets for VWN. For experiment and evaluation, a sub VWN in IT area was generated by an implemented tool that takes input is a list of IT keywords. We compare the outputs with existing Vietnamese resources. The results show that 86.6% of synsets generated by our method is equivalent to synsets created by human.