Data Fusion Scheme Based on the Intersection of Privacy Sets

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

In the field of public service data integration, government agencies are responsible for consolidating data from various industries and establishing big data platforms. Some achievements have been made in this direction, but there are still some issues related to privacy and security. In this process, determining how a government agency can identify the “common data category” between different platforms without directly viewing the private data of each platform is a challenge. To address this issue, we propose a three-party data fusion method based on the intersection of marked privacy sets, in which two platforms privately compute their common data categories while a government agency acts as a keyless comparer. For the traditional privacy set intersection protocol, we designed a set of matrix mapping methods. By mapping the set elements to the matrix, we avoid direct access to the data. This protocol can confirm the “common data category” between the participants without exposing the private elements, thereby achieving the purpose of privacy protection. Experimental results show that our protocol can calculate the “common data category” of each platform under appropriate communication and computational volumes.