Privacy-preserving recommender system using the data collaboration analysis for distributed datasets

TL;DR

This paper proposes a privacy-preserving framework for recommender systems that enables multiple parties to collaboratively improve prediction accuracy without compromising personal data security.

Contribution

It introduces a novel data collaboration analysis method that enhances rating prediction accuracy while maintaining privacy across distributed datasets.

Findings

Improved prediction accuracy in distributed datasets

Effective privacy preservation in recommender systems

Potential for broader application in privacy-sensitive domains

Abstract

In order to provide high-quality recommendations for users, it is desirable to share and integrate multiple datasets held by different parties. However, when sharing such distributed datasets, we need to protect personal and confidential information contained in the datasets. To this end, we establish a framework for privacy-preserving recommender systems using the data collaboration analysis of distributed datasets. Numerical experiments with two public rating datasets demonstrate that our privacy-preserving method for rating prediction can improve the prediction accuracy for distributed datasets. This study opens up new possibilities for privacy-preserving techniques in recommender systems.