A blockchain-based trading system for big data

Data are an extremely important asset. Governments around the world encourage big data sharing and trading to promote the big data economy. However, existing data trading platforms are not fully trusted. Such platforms face the problems of a single point of failure (SPOF), opaque transactions, uncontrollability, untraceability, and issues of data privacy. Several blockchain-based big data trading methods have been proposed; however, they do not adequately address the security issues introduced by dishonesty in the data provider and data agent or the fairness of data revenue distribution and price bargaining. In this paper, we propose a blockchainbased decentralized data trading system in which data trading is completed by smart contract-based data matching, price negotiation, and reward assigning. Moreover, the proposed data trading system evaluates the data quality on the basis of three metrics, records the evaluation results in a side-chain, and distributes the data users' application revenue to the data provider according to the evaluated data quality. We verify the security, usability, and efficiency of the proposed big data trading system.