Learning to Predict the Quality of Contributions to strongWikipediastrong.pdfVIP

Learning to Predict the Quality of Contributions to Wikipedia Gregory Druck and Gerome Miklau and Andrew McCallum {gdruck,miklau,mccallum}@ Department of Computer Science University of Massachusetts Amherst, MA 01003 Abstract in debates on article discussion pages. These users are sometimes referred to as “zealots” (Anthony, Smith, and Although some have argued that Wikipedia’s open edit pol- Williamson 2007), and studies claim that they are moti- icy is one of the primary reasons for its success, it also raises concerns about quality — vandalism, bias, and errors can be vated by a system of peer recognition that bears resemblance problems. Despite these challenges, Wikipedia articles are to the academic community (Forte and Bruckman 2005). often (perhaps surprisingly) of high quality, which many at- However, the contributions of “good Samaritan” users, who tribute to both the dedicated Wikipedia community and “good edit articles but have no desire to participate in the com- Samaritan” users. As Wikipedia continues to grow, however, munity, cannot not be underestimated (Anthony, Smith, and it becomes more difficult for these users to keep up with the Williamson 2007). increasing number of articles and edits. This motivates the As Wikipedia continues to