View article

With the demand of bilingual dictionaries covering domain-specific terminology, research in the field of automatic dictionary extraction has become popular. However, present research based on the analysis of bilingual text corpora faces several issues regarding very different language pairs such as Japanese and English. Therefore, we present an approach to extracting an English-Japanese dictionary from the link structure of Wikipedia, a huge scale multilingual Web-based encyclopedia. We extracted a large amount of translation pairs and conducted some experiments in which we compared the accuracy and coverage to a dictionary trained on a parallel corpus.