Extending WordNet with Hypernyms and Siblings Acquired from Wikipedia
Ichiro Yamada, Jong–Hoon Oh, Chikara Hashimoto, Kentaro Torisawa, Jun’ichi Kazama, Stijn De Saeger, Takuya Kawada · International Joint Conference on Natural Language Processing · 2011
This paper proposes a method for extending WordNet with terms in Wikipedia. Our method identifies a WordNet synset by integrating evidence derived from the structure of an article in Wikipedia and distributional similarity of terms. Unlike previous methods, utilizing the hypernym and siblings of the target term acquired from Wikipedia, the proposed method can deal with terms other than Wikipedia article titles and can work well even when reliable distributional similarity of a target term is unavailable. Experiments show that the proposed method can identify synsets for 2,039,417 inputs at precision rate of 84%. Furthermore, it is estimated from the experimental results that there should be 328,572 terms among all the inputs whose synset our method can correctly identify, while previous methods relying only on distributional similarity and lexico-syntactic patterns cannot.