Skip to main content
Article
Wikipedia Vandalism Detection: Combining Natural Language, Metadata, and Reputation Features
Lecture Notes in Computer Science: Computational Linguistics and Intelligent Text Processing
  • B. Thomas Adler, University of California, Santa Cruz
  • Luca de Alfaro, University of California, Santa Cruz -- Google
  • Santiago M Mola-Velasco, Universidad Politcnica de Valencia
  • Paolo Rosso, Universidad Politcnica de Valencia
  • Andrew G. West, University of Pennsylvania
Date of this Version
2-1-2011
Document Type
Conference Paper
Comments
CICLing '11: Proceedings of the 12th International Conference on Intelligent Text Processing and Computational Linguistics, Tokyo, Japan, February 20-26, 2011.
Abstract

Wikipedia is an online encyclopedia which anyone can edit. While most edits are constructive, about 7% are acts of vandalism. Such behavior is characterized by modifications made in bad faith; introducing spam and other inappropriate content. In this work, we present the results of an effort to integrate three of the leading approaches to Wikipedia vandalism detection: a spatio-temporal analysis of metadata (STiki), a reputation-based system (WikiTrust), and natural language processing features. The performance of the resulting joint system improves the state-of-the-art from all previous methods and establishes a new baseline for Wikipedia vandalism detection. We examine in detail the contribution of the three approaches, both for the task of discovering fresh vandalism, and for the task of locating vandalism in the complete set of Wikipedia revisions.

DOI
10.1007/978-3-642-19437-5_23
Copyright/Permission Statement
The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-19437-5_23
Keywords
  • Wikipedia,
  • wiki,
  • collaboration,
  • vandalism,
  • machine learning,
  • metadata,
  • natural-language processing,
  • reputation
Citation Information
B. Thomas Adler, Luca de Alfaro, Santiago M Mola-Velasco, Paolo Rosso, et al.. "Wikipedia Vandalism Detection: Combining Natural Language, Metadata, and Reputation Features" Lecture Notes in Computer Science: Computational Linguistics and Intelligent Text Processing Vol. 6609 (2011) p. 277 - 288
Available at: http://works.bepress.com/andrew_g_west/9/