Japanese Case Structure Analysis by Unsupervised Construction of a Case Frame Dictionary.pdfVIP

Japanese Case Structure Analysis by Unsupervised Construction of a Case Frame Dictionary Daisuke Kawahara, Nobuhiro Kaji and Sadao Kurohashi Graduate School of Informatics, Kyoto University Yoshida-Honmachi, Sakyo-ku, Kyoto, 606–8501, Japan {kawahara,kaji,kuro}@pine.kuee.kyoto-u.ac.jp Abstract tically analyzed corpora are too small to learn a dictionary, since case frame information consists In Japanese, case structure analysis is very im- of relations between nouns and verbs, which mul- portant to handle several troublesome charac- tiplies to millions of combinations. teristics of Japanese such as scrambling, omis- Based on such a consideration, we took the sion of case components, and disappearance of following unsupervised learning strategy to the case markers. However, for lack of a wide- Japanese case structure analysis: coverage case frame dictionary, it has been dif- ficult to perform case structure analysis accu- 1. At first, a robust and accurate parser is de- rately. Although several methods to construct veloped, which does not utilize a case frame a case frame dictionary from analyzed corpora dictionary, have been proposed, they cannot a