Utilizing Features of Verbs in Statistical Zero Pronoun Resolution for Japanese Speech
Sen Yoshida, Masaaki Nagata · Waseda University Repository (Waseda University) · 2009
This paper proposes a statistical zero pronoun resolution method that utilizes features of verbs.In Japanese speech, the subject is often omitted, especially when it is the first person.To resolve such zero pronouns, features related to the verbs such as functional expressions play important roles.However, recent state-of-the-art zero-pronoun resolution systems lack these features because they are mainly designed for written texts such as newspaper articles, in which first person subjects are rare.We show that a set of verbal features has the ability to distinguish first persons from others in monologue transcriptions, and this improves the accuracy of zero pronoun resolution with statistical machine learning.