Calculating and Presenting Trust in Collaborative Content

Collaborative functionality is increasingly prevalent in Internet applications. Such functionality permits individuals to add -- and sometimes modify -- web content, often with minimal barriers to entry. Ideally, large bodies of knowledge can be amassed and shared in this manner. However, such software also provides a medium for biased individuals, spammers, and nefarious persons to operate. By computing trust/reputation for participating agents and/or the content they generate, one can identify quality contributions. In this work, we survey the state-of-the-art for calculating trust in collaborative content. In particular, we examine four proposals from literature based on: (1) content persistence, (2) natural-language processing, (3) metadata properties, and (4) incoming link quantity. Though each technique can be applied broadly, Wikipedia provides a focal point for discussion. Finally, having critiqued how trust values are calculated, we analyze how the presentation of these values can benefit end-users and application security.