Đang chuẩn bị nút TẢI XUỐNG, xin hãy chờ
Tải xuống
For biomedical information extraction, most systems use syntactic patterns on verbs (anchor verbs ) and their arguments. Anchor verbs can be selected by focusing on their arguments. We propose to use predicate-argument structures (PASs), which are outputs of a full parser, to obtain verbs and their arguments. In this paper, we evaluated PAS method by comparing it to a method using part of speech (POSs) pattern matching. POS patterns produced larger results with incorrect arguments, and the results will cause adverse effects on a phase selecting appropriate verbs. . | Finding Anchor Verbs for Biomedical IE Using Predicate-Argument Structures Akane YAKUSHIJIt Yuka TATEISItt YusukeMIYAOt Jun ichi TSUJIItt fDepartment of Computer Science University of Tokyo Hongo 7-3-1 Bunkyo-ku Tokyo 113-0033 JAPAN tCREST JST Japan Science and Technology Agency Honcho 4-1-8 Kawaguchi-shi Saitama 332-0012 JAPAN akane yucca yusuke tsujii @is.s.u-tokyo.ac.jp Abstract For biomedical information extraction most systems use syntactic patterns on verbs anchor verbs and their arguments. Anchor verbs can be selected by focusing on their arguments. We propose to use predicate-argument structures PASs which are outputs of a full parser to obtain verbs and their arguments. In this paper we evaluated PAS method by comparing it to a method using part of speech POSs pattern matching. POS patterns produced larger results with incorrect arguments and the results will cause adverse effects on a phase selecting appropriate verbs. 1 Introduction Research in molecular-biology field is discovering enormous amount of new facts and thus there is an increasing need for information extraction IE technology to support database building and to find novel knowledge in online journals. To implement IE systems we need to construct extraction rules i.e. rules to extract desired information from processed resource. One subtask of the construction is defining a set of anchor verbs which express realization of desired information in natural language text. In this paper we propose a novel method of finding anchor verbs extracting anchor verbs from predicate-argument structures PASs obtained by full parsing. We here discuss only finding anchor verbs although our final purpose is construction of extraction rules. Most anchor verbs take topical nouns i.e. nouns describing target entities for IE as their arguments. Thus verbs which take topical nouns can be candidates for anchor verbs. Our method collects anchor verb candidates by choosing PASs whose arguments are topical nouns. Then .
