Using Syntactic Structures and Cohesive Devices in Recognizing Discourse Structure of Text(Text Mining I)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)

スポンサーリンク

概要

• 論文の詳細を見る

• This paper presents a system for automatically generating the discourse structure of text. The system is divided into two levels: sentence-level and text-level. At the sentence-level, the discourse analyzer uses syntactic structures and cue phrases to derive discourse structures of sentences. This approach prevents combinatorial explosions while still get accurate analyses. At the text-level, constraints about textual adjacency and textual organization are integrated in a beam search to reduce the search space of the discourse analyzer and generate the best discourse structure. Two new factors using to recognize discourse relations are proposed: noun-phrase cues and verb-phrase cues. Our experiments show that this system achieves a good performance compared to existing discourse analyzing systems.

• 一般社団法人電子情報通信学会の論文
• 2004-11-30

著者

• Abeysinghe Geetha

  School Of Computing Science Middlesex University

• Thanh Huong

  Faculty Of Information Technology Hanoi University Of Technology

• Lethanh Huong

  Faculty Of Information Technology Hanoi University Of Technology

関連論文

• Using Syntactic Structures and Cohesive Devices in Recognizing Discourse Structure of Text(Text Mining II)
• Using Syntactic Structures and Cohesive Devices in Recognizing Discourse Structure of Text(Text Mining I)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)

スポンサーリンク