A dual-guarantee approach for cross-departmental government data sharing with usable-but-invisible privacy based on privacy computing and blockchain

Abstract

Cross‑departmental government data sharing is essential for integrated public services, yet it persistently faces tension between data utility and privacy protection. Existing approaches either enforce rigid access control that hampers analytical depth or apply heavy‑duty encryption that degrades query performance. This paper proposes a dual‑guarantee sharing framework that combines privacy computing and blockchain to achieve usable‑but‑invisible data circulation. The privacy computing layer employs a hybrid of secure multiparty computation and differential privacy mechanisms to enable joint statistical analysis without exposing raw records. The blockchain layer ensures traceable record keeping, attribute authentication and audit trail of data. A simple evaluation method estimates the usage of privacy budget, the decrease of data utility and the transaction speed. The framework is tested on actual administrative data sets which include tax, social security and housing fund records from three municipal departments. Experimental outcomes indicate that the system maintains an analytical accuracy within 4% of the plain text ones while keeping the total privacy budget less than 0.8, can process more than 120 simultaneous sharing requests per second and completes the consensus of the blockchain within 1.2 seconds. A pilot study in a provincial government information exchange platform shows the easy combination with the present e-government facilities and shortens the time of inter-departmental data retrieval from 5 working days to less than 3 minutes. This method offers a feasible and legally acceptable plan for confidential governmental data cooperation.