图模型方法的演化式摘要研究  

An Exploration of Graph-Based Evolutionary Summarization

作　　者：贾永辉 王捷 王超 沈佳斌 袁涛 王红玲[1] JIA Yonghui;WANG Jie;WANG Chao;SHEN Jiabin;YUAN Tao;WANG Hongling(Inst. Of Computer Science and Technology, Soochow University, Suzhou, China, 215006)

机构地区：[1]苏州大学计算机科学与技术学院

出　　处：《福建电脑》2019年第5期16-20,共5页Journal of Fujian Computer

基　　金：苏州大学“大学生创新创业训练计划”(No.201710285092X)资助

摘　　要：面向互联网新闻事件的演化式摘要是自然语言处理的一个新兴任务,其本质是多文档自动文摘。由于互联网新闻事件报道具有动态演化、内容关联和信息重复等特点,面向互联网新闻事件的演化式文摘与传统文摘相比存在诸多不同。本文针对该任务,提出了一种演化式摘要的方法,包括事件阶段划分方法和使用基于图模型的方法对演化式摘要进行建模。在实验过程中,本文选择了一个具有完整的发展过程的热点事件为实验对象,实验结果表明,与传统方法相比,使用基于图模型的方法取得了良好的性能。Evolutionary summarization for Internet News Event is a new task in natural language processing,which is essentially multi-document summarization. Because of the characteristics of dynamic evolution, content relation and information redundancy of Internet news reports, there are many differences between the evolutional summarization and traditional summarization. For this task, this paper proposes a method for evolutionary summarization, which includes dividing the event development stages and using the graph model-based method to model the evolutionary summary. In the experimental process, this paper selects a high-impact event with a complete development process as the experimental object. The experimental results show that compared with the traditional method, the graph-based model method achieves better performance.

关 键 词：事件阶段划分 图模型 演化式摘要 PAGERANK 多文档自动文摘 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]