Research on Data Sharing Model Based on Multi-party Secure Computing
Yu Zhang, Lili Yang, Wenxi Long, Jun Han · 2022
In order to better solve the problems of industry cooperation, mutual trust and data sharing, more accurate detection and prevention of security problems hidden in the business data of the State Grid, to achieve cross-industry model sharing training and ecological construction. This research proposes a multi-party data sharing model training engine based on "blockchain + federated learning", which can be used in scenarios such as smart retail, risk assessment and satisfaction prediction to achieve multi-party privacy data sharing, build a data ecology, break data silos, and mine data. Joint value, enabling multi-party secure computing. After testing, it is found that compared with the traditional consensus mechanism, the information sharing encryption scheme proposed in the article is more complete and more efficient in encryption processing, and can be used as a shared information encryption processing tool.