Designing Socially-Optimal Rating Protocols for Crowdsourcing Contest Dilemma
Jianfeng Lu,Changbing Tang,Xiang Li,Qian Wu
DOI: https://doi.org/10.1109/tifs.2017.2656468
IF: 7.231
2017-01-01
IEEE Transactions on Information Forensics and Security
Abstract:Despite the increasing popularity and the perceived promise of crowdsourcing, its openness presents individuals with an opportunity to exhibit antisocial behavior, such as free-ride and attack to decrease the social welfare, which is considered as a crowdsourcing contest dilemma. Hence, incentive mechanisms are needed to compel rational and selfish individuals to contribute well behavior in tasks. In this paper, we integrate the pricing and reputation schemes to design a novel socially optimal rating protocol based on game theory, in which each player is tagged with a rating to represent its social status, and players are encouraged to contribute good behaviors to increase their ratings, thus receive higher rewards. In particular, we analyze how the players' behaviors are influenced by the incurred costs and the designed payment, as well as their long-term utilities. By quantifying the sufficient and necessary conditions under which all players comply with the social norm in their self-interests, we formulate the rating protocol design problem, and analyze the impacts of the design parameters in order to characterize the optimal design, that maximizes the social welfare to achieve the social optimum. Finally, illustrative results show the validity and effectiveness of our proposed protocol design for crowdsourcing contest dilemma.