• DocumentCode
    1532096
  • Title

    The InfoFinder agent: learning user interests through heuristic phrase extraction

  • Author

    Krulwich, Bruce ; Burkey, Chad

  • Author_Institution
    AgentSoft Ltd., Jerusalem, Israel
  • Volume
    12
  • Issue
    5
  • fYear
    1997
  • Firstpage
    22
  • Lastpage
    27
  • Abstract
    The paper discusses InfoFinder, a system that uses several innovative approaches for learning user information interests from sets of messages or other online documents that users have classified. InfoFinder´s innovations leverage the unique aspects of learning from user-classified documents in a real-world setting. Users find Infofinder to be an accurate and valuable tool for finding information relating to ongoing interests
  • Keywords
    heuristic programming; information needs; information retrieval; learning (artificial intelligence); software agents; InfoFinder; heuristic phrase extraction; information retrieval; innovative approaches; learning; messages; online documents; software agent; user interests; user-classified documents; Data mining; Explosives; Internet; Java; Machine learning; Maintenance; Search engines; Search methods; Technological innovation; Web sites;
  • fLanguage
    English
  • Journal_Title
    IEEE Expert
  • Publisher
    ieee
  • ISSN
    0885-9000
  • Type

    jour

  • DOI
    10.1109/64.621224
  • Filename
    621224