Philgoo Han writeup of Giles Talk

From Cohen Courses
Jump to navigationJump to search
  • SeerSuite: Infrastucture for research in IR + IE
    • First commercial robust infrastructure
    • Broad area of usage
  • Bringing small science (in the sense of shared pool of information) into the big
  • CiteSeer-x
    • Extract information from pdf files, no use of metadata
    • Crawling the web: backtrack user to there academic homepages
    • More utilizing user reaction(?) knowledge may be interesting
    • Some improvements may be introduced in entity disambiguation(somethings from class may)