DARS: Empowering Trust in Blockchain-Based Real-World Applications with a Decentralized Anonymous Reputation System

Abstract

Blockchain-based Reputation Systems (BRS) are a recent and essential development in decentralized trust and reputation management. The decentralization, transparency, and efficiency brought by Blockchain (BC) are clearly what we always hoped for to build effective trustless reputation systems. Despite these promising attributes, existing BRS face a critical challenge in countering common reputation attacks, including whitewashing, self-promotion, and bad-collision attacks. Currently, BRS rely on reputation scores or tokens linked to the same address or public key, which severely limits their widespread adoption as it raises concerns about possible retaliation, hence the reluctance of users to engage and provide feedback. In this work, we propose and develop a Decentralized Anonymous Reputation System (DARS) for trust-related applications. In DARS, users can use different pseudonyms when interacting with each other to hide their digital identities. In our system design, all pseudonyms of a specific user, yet, are cryptographically linked to the same access token, allowing honest users to maintain their reputation and preventing malicious ones from starting over. This is achieved through the use of zkSNARK proofs for set membership via Merkle trees over commitments. We extended our framework with an efficient reputation model that respects all the security and privacy properties of our formal model. Finally, we developed a prototype of the proposed framework using emerging technologies and cryptographic tools. The evaluation results demonstrate the feasibility and effectiveness of DARS.