Skip to main navigation Skip to search Skip to main content

Research on dynamic multi-document summarization by topic detection and tracking technology

  • School of Computer Science and Technology, Harbin Institute of Technology
  • Northeast Forestry University

Research output: Contribution to journalArticlepeer-review

Abstract

In this paper time characteristics of dynamic multi-documents summarization has been analyzed by the clustering idea in topic detection technology. From the change of time information threshold value, different multi-document clustering and multi-documents summarization for the dynamic Web information data stream can be generated. By compared with different threshold values, the importance of time information in a dynamic multi-document summarization is understanded. Experimental result means that you can counterpoise the output of TDT by adjusting α value and generate better quality dynamic multi-document summarization.

Original languageEnglish
Pages (from-to)1767-1770
Number of pages4
JournalHarbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology
Volume42
Issue number11
StatePublished - Nov 2010
Externally publishedYes

Keywords

  • Dynamic threshold
  • Multi-document summarization
  • Temporal properties
  • Topic detection

Fingerprint

Dive into the research topics of 'Research on dynamic multi-document summarization by topic detection and tracking technology'. Together they form a unique fingerprint.

Cite this