A Novel Reputation Management Mechanism with Forgiveness in P2P File Sharing Networks

Abstract

In peer-to-peer (P2P) file sharing networks, it is common practice to manage each peer using reputation systems. A reputation system systematically tracks the reputation of each peer and punishes peers for malicious behaviors (like uploading bad file, or virus, etc). However, current reputation systems could hurt the normal peers, since they might occasionally make mistakes. Therefore, in this paper, we introduce forgiveness mechanism into the EigenTrust reputation system to reduce such malicious treatments and give them opportunities to gain reputation back. Particularly, we take four motivations (the severity of current offence, the frequency of offences, the compensation and the reciprocity of the offender) into consideration to measure forgiveness. The simulation work shows that the forgiveness model can repair the direct trust breakdown caused by unintentional mistakes and lead to less invalid downloads, which improves the performance of P2P file sharing systems.