语篇结构标注钻研的综述(4)

    七.结论
    二00一年,由Daniel Marcu博士主持的钻研小组以RST理论为支持创建了语篇标注语料库。钻研小组所标注的三八五篇华尔街报文章皆取自宾州树库,篇幅长度不等,从三一个词到二,一二四个词,总词数到达一七六,000,均匀每一篇文章四五八个词。文章的内容触及到各种话题,如财政报道、贸易新闻、文化点评、编者按、读者来信等。语料库建设的主要成绩为:确立了如何将语篇切分为基本语篇单位的理论、扩铺了修辞瓜葛集、为RST理论的应用提供了广阔的遥景。
    参考文献:
    [一] [ZK(#]Carlson,L.,Marcu.D.& Okurowski M.Building a Discourse_tagged Corpus in the Framework of Rhetorical Structure Theory.Proceedings of the First Annual Meeting of the North American Chapter of the Association for Computational Linguistics,Seattle,WA,二00一:九-一七.
    [二] Grosz,B.& Sidner,C.Attentions,Intentions,and the Structure of Discourse[J].?Computational Linguistics?,一二(三):一七五-二0四.Talmy Givon,一九八三/一九八六.
    [三] Halliday,M.A.K.& R.Hasan.?Cohesion in English?[M].London:Longman,一九七六.
    [四] Mann.W.& S.Thompson.Rhetorical Structure Theory:A Theory of Text Organization.USC Information Science Institute.Technical Report I (SI/ RS-八七-一九0),一九八七.
    [五] Marcu,D.?The Theory and Practice of Discourse Parsing and Su妹妹arization?[M].Cambridge,Massachusetts:MIT Press,二000.