Predicate-Argument Structure Analysis and Zero-Anaphora Resolution for Dialogue Analysis

Abstract

This paper presents a predicate-argument structure (PAS) analysis for dialogue systems in Japanese. Conventional PAS analyses have been applied to newspaper articles; however, there are differences between newspapers and dialogues. Therefore, to comprehensively deal with these differences, we constructed a PAS analyzer for dialogues as a type of domain adaptation from newspapers. Because pronominalization and ellipses frequently appear in dialogues, we utilized a strategy that simultaneously resolves zero-anaphora and adapts our PAS analyzer to dialogues. By incorporating parameter adaptation and automatically acquiring knowledge from large text corpora, we performed a PAS analysis that is specific to dialogues. Our PAS analyzer has a higher accuracy compared with the analyzer for newspaper articles.

Domain Adaptation in Japanese Predicate-Argument Structure Analysis considering First and Second Person Exophora

“…none (Zhou and Xu 2015;He, Lee, Lewis, and Zettlemoyer 2017) 1 null-subject (Iida and Poesio 2011;Rello, Baeza-Yates, and Mitkov 2012;Chen and Ng 2016;Yin, Zhang, Zhang, and Liu 2017) (Matsubayashi and Inui 2017;Hayashibe, Komachi, and Matsumoto 2011;Imamura, Higashinaka, and Izumi 2014;Shibata, Kawahara, and Kurohashi 2016;Ouchi, Shindo, Duh, and Matsumoto 2015;Yoshikawa, Asahara, and Matsumoto 2013;Taira, Fujita, and Nagata 2008) end-to-end (Ouchi, Shindo, and Matsumoto 2017;Matsubayashi and Inui 2018) (1) Fine-tuning (FT) 2…”