DSRep: Implementing decentralized reputation evaluation for data space based on a low-cost and high-throughput blockchain

Data spaces are emerging as a promising solution for secure and decentralized data sharing across organizations. However, in such open environments, it is difficult for participants to establish trust in the quality of data products and the behavior of users. To address this challenge, we propose DSRep, a dynamic and decentralized reputation evaluation model tailored for data spaces. Based on a survey of 31 real-world data exchange platforms, DSRep introduces role-specific evaluation metrics for data products, data providers, and data consumers. DSRep proposes a DAO-based governance mechanism allows dynamic configuration of metrics and weights through on-chain voting. To handle the high-frequency reputation logs efficiently, DSRep propose a blockchain processing mechanism with two-level transaction pools. In the experiments, DSRep can effectively evaluate the reputation of users and data products, while reducing storage consumption by 89.97% and improving throughput by 117.28%.