Building a Reputation System for Wikipedia

Wikipedia is a web-based encyclopedia, written and edited collaboratively by Internet users. Wikipedia has an extremely open editorial policy that allows anybody to create or modify articles. This has promoted a broad and detailed coverage of subjects, but also introduced problems relating to the quality of articles.

The Wikipedia Recommender System (WRS) was developed to help users determine the credibility of articles based on feedback from other Wikipedia users. The WRS provides a rating which emphasizes feedback from recommenders that the user has agreed with in the past. This paper presents some of the work that has gone into the development of the Wikipedia Recommender System.

We first developed a generic architecture for integrating a reputation system into existing legacy systems and based our design of the WRS on this architecture. Both the generic architecture and our design of the WRS are outlined in this paper. Finally, we present ongoing work to improve the reputation rating of the WRS by determining the areas of expertise for the different feedback providers in the WRS.

This will allow more accurate recommendations because the system can assign a higher weight to feedback from recommenders that have previously demonstrated competence in the area of the article. In order to determine the areas of expertise of recommenders, however, we first need to identify a way to classify content in Wikipedia.

We outline current efforts to evaluate different classification schemes and illustrate how knowing the expertise of recommenders may help us when we calculate the rating for a Wikipedia article.