Japanese Ellipsis Resolution in "A NO B" Noun Phrases for Colloquial Inquiry Text Using Latent Topic Models
Tomohiko Harada, Nobuo Suzuki, Kazuhiko Tsuda · 2013
Generally inquiries through Web forms and e-mails are increasing. These inquiry texts usually include many informal expressions use of the colloquial style, such as a spoken language, and many omitted words. An omitted word causes the meaning of a sentence to become ambiguous and may make the reader misread and misunderstand the context. In this paper we focus on the frequently omitted noun ``B'' in the noun phrase ``A NO1 B'' (usually meaning B of A) seen in the colloquial style inquiry text and propose a method to predict omitted noun ``B'' from context and knowledge using topic information. From the results of the evaluation experiment, we have confirmed that our method improved 11.34 points from the conventional method, and predicted the omitted word with an accuracy rate of more than 75% using ``Latent Dirichlet Allocation'' (LDA.)