Chinese Journal of Computers   Full Text
Title	Research on Dynamic Summarization for Evolutionary Web Information
Authors	ZHANG Jin1),2) XU Hong-Bo1) CHENG Xue-Qi1)
Address	1)(Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100080) 2)(Graduate University of Chinese Academy of Sciences, Beijing 100039)
Year	2008
Issue	No.4(696—701)
Abstract &   Background	Abstract With the development of Internet and the emergence of Web2.0, the dynamic evolution of Web information is becoming more and more evident. This paper presents the formalization of dynamic summarization according to temporal evolution of Web information. Based on the analysis of evolution of Web information, the authors measure the difference between current information and history information, and then propose three models for dynamic summarization and the solutions in detail. Experiments on DUC 2007 update dataset illustrate the performance of the methods. keywords dynamic evolution； dynamic summarization； model； performance background The authors are members of Web mining and search group of Institute of Computing Technology, Chinese Academy of Sciences. The major research interests of this group are shallow natural language processing, text mining and information retrieval. The work is supported by the National Basic Research Program(973 Program) of China "Large-Scale Text Content Computing" under grand No.2004CB318109. With the development of Internet and the emergence of Web2.0, dynamic summarization becomes an important research question especially with voluminous coverage of a long-running story. This paper presents the definition of dynamic summarization according to temporal analysis and then proposes the fundamental content filtering models for the identification of dynamic information. Furthermore, the solutions for the content filtering models are proposed.