• DocumentCode
    2227761
  • Title

    Learning to perform moderation in online forums

  • Author

    Arnt, Andrew ; Zilberstein, Shlomo

  • Author_Institution
    Dept. of Comput. Sci., Massachusetts Univ., Amherst, MA, USA
  • fYear
    2003
  • fDate
    13-17 Oct. 2003
  • Firstpage
    637
  • Lastpage
    641
  • Abstract
    Online discussion forums are a valuable resource for people looking to find information, discuss ideas, and get advice on the Internet. Unfortunately, many forums have too much activity and information available, resulting in information overload. Moderation systems are implemented in some forums as a way to handle this problem, but due to sparsity issues, they are often not sufficient. We describe a novel method for learning from past moderations to develop a classifier that can perform automated moderation and thus address the sparsity problem. Additionally, we discuss the possibility of training a moderating classifier on a moderated forum and then applying it to an otherwise unmoderated forum.
  • Keywords
    Internet; classification; information filters; information retrieval; learning (artificial intelligence); Internet; information overload; machine learning; moderated forum; moderating classifier; moderation systems; online discussion forums; sparsity issues; unmoderated forum; Accuracy; Aggregates; Collaboration; Computer science; Discussion forums; Filtering algorithms; Filters; Humans; Recommender systems; Search engines;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Web Intelligence, 2003. WI 2003. Proceedings. IEEE/WIC International Conference on
  • Print_ISBN
    0-7695-1932-6
  • Type

    conf

  • DOI
    10.1109/WI.2003.1241285
  • Filename
    1241285