The most straightforward method to rate users and services in a sharing economy is to consider their average ratings (we refer it as the mean method). However, such methods are very sensitive to the noisy information and manipulation. In these rating systems, some users may give unreasonable ratings because they are not serious about the rating or simply not familiar with the related field. In addition, the system may contain some malicious spammers who always deliberately give high ratings to some low quality participants. To solve this problem, rating algorithms to prevent spamming are proposed. These algorithms build a reputation system for users. The ratings of users with higher reputation are assigned with more weight. By iteratively updating users’ reputation, the reputation of other users and services can be ranked more accurately than the average ratings method.
[[ Diagram ]]
We’ve used a reputation redistribution process to the iterative ranking algorithm, which can effectively enhance the weight of the highly reputed users and lower the weight of the users with low reputation in estimating the quality of objects. The tests of this method in both artificial and real data show that the accuracy of objects’ quality ranking is considerably improved. Moreover, two penalty factors to the iterative ranking algorithm significantly improve its robustness against the malicious spamming behaviors. [https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0097146&type=printable]