基于信息融合的多文档自动文摘技术  

Multi-Document Automatic Summarization Technique Based on Information Fusion

作　　者：徐永东[1] 徐志明[1] 王晓龙[1] 

机构地区：[1]哈尔滨工业大学计算机学院智能技术与自然语言处理研究室,哈尔滨150001

出　　处：《计算机学报》2007年第11期2048-2054,共7页Chinese Journal of Computers

基　　金：国家自然科学基金重点项目(60435020);国家自然科学基金(60773070)资助.~~

摘　　要：提出了一个面向多文档自动文摘任务的多文本框架(Multiple Document Framework,MDF),该框架通过系统地描述不同层面的文本单元之间的相互关系以及文档集合蕴含的事件在时间上的发生及演变,将多篇文档在不损失文档集合原有信息的前提下实现信息融合.MDF简化了传统交叉文本结构理论的文本集合表示模型,又补充了信息融合理论中缺乏的事件主题的演变性和分布性信息.文中给出了建立MDF、基于MDF的信息融合、文摘生成等一整套算法.通过对32组不同主题的网络文档试验结果表明,MDF策略很好地实现了多知识源的并行融合,并获得了较好的结果.A Multiple Documents Framework （MDF） is proposed for multi-document automatic summarization task. By representing interrelationship between text units at different levels of granularity and the happen and change of various events at time dimension, this framework can achieve information fusion of multi-document while reserve original information of set of related documents. MDF simplifies traditional multi-document representation in cross structure theory and simultaneously, supplements change and distribution informations of events topics which cannot be obtained in information fusion theory. Concretely, a series of algorithms including building MDF, multi-document information fusion based MDF and summarization generation are proposed. The capability of concurrently fusing multiple knowledge sources of MDF strategies is testified by experiments in 32 different sets of net documents and shows good results.

关 键 词：多文本框架 多文档自动文摘 信息融合 时间 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]