An Environment for Extracting Resolution Rules of Zero Pronouns from Corpora

要旨

This paper proposes a practical integrated environment for extracting rules for the anaphora resolution of zero pronouns from monolingual and/or bilingual corpora. This method takes into account the practical situation for making resolution rules of zero pronouns in specific domain texts; the types of usable corpora (monolingual and/or bilingual) for examining the extraction of resolution rules have been changed depending on the type of NLP system using extracted resolution rules. The extraction processes of resolution rules in the environment are classified into five component tasks: (1) Zero Pronoun Identification, (2) Antecedent Annotation, (3) Rejection of Sentences Unsuitable for Rule Extraction, (4) Rule Extraction, and (5) Extracted Rule Application and Modification. An automatic process and/or a manual process with a user friendly human interface can be used to achieve each component task. This environment was implemented in the Japanese-to-English machine translation system, ALT-J/E, for Japanese zero pronoun resolution.

主題