• DocumentCode
    480137
  • Title

    Improving the Performance of Topic Tracking System by Ensemble

  • Author

    Qin, Xiangju ; Zhang, Yang

  • Author_Institution
    Coll. of Inf. Eng., Northwest A&F Univ., Yangling
  • Volume
    4
  • fYear
    2008
  • fDate
    12-14 Dec. 2008
  • Firstpage
    316
  • Lastpage
    320
  • Abstract
    The aim of topic tracking is to monitor the stream of news stories to find additional stories on a topic that was identified by several sample stories. In this paper, we adopt majority voting to ensemble several topic tracking systems. The experiment results on TDT3 corpus show that our ensemble topic tracking system performs better than each individual topic tracking system, and that the performance of topic tracking system could be improved by ensemble technology.
  • Keywords
    pattern classification; text analysis; binary classification problem; ensemble technology; ensemble topic tracking system; majority voting; text classification; Classification algorithms; Computer science; Educational institutions; Information filtering; Information filters; Information retrieval; Monitoring; Performance evaluation; Software engineering; Voting; Ensemble; TDT3 Corpus; Topic Tracking;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Software Engineering, 2008 International Conference on
  • Conference_Location
    Wuhan, Hubei
  • Print_ISBN
    978-0-7695-3336-0
  • Type

    conf

  • DOI
    10.1109/CSSE.2008.984
  • Filename
    4722624