Enhancing User Privacy in Personalized Recommendation Systems
Rongsheng Cai, Suzhen Luo · 2024
In the data-driven era, recommendation systems have become deeply integrated into our daily lives, helping users discover products, content, or services of interest. However, these systems often handle a vast amount of sensitive data, such as user interests, behaviors, and social information. Balancing personalized service with privacy protection is a significant challenge in the design of current recommendation systems. Private Set Intersection (PSI), which is an important privacy-preserving protocol, computes an intersection of private datasets. For the recommendation system, we introduces an innovative approach by applying PSI to recommendation systems, enabling efficient data analysis and personalized recommendations while safeguarding user privacy.