Algorithms and incentives for robust ranking
Rajat Kanti Bhattacharjee, Ashish Goel · 2007
Spam in the form of link spam and click spam has become a major obstacle in the effective functioning of ranking and reputation systems. Even in the absence of spam, difficulty in eliciting feedback and self-reinforcing nature of ranking systems are known problems. In this paper, we make a case for sharing with users the revenue generated by such systems as incentive to provide useful feedback and present an incentive based ranking scheme in a realistic model of user behavior which addresses the above problems. We give an explicit ranking algorithm based on user feedback. Our incentive structure and ranking algorithm ensure that there is a profitable arbitrage opportunity for the users of the system in correcting the inaccuracies of the ranking. The system is oblivious to the source of inaccuracies (benign or