A Practical Privacy-Preserving Recommender System

• DOI: 10.1007/s41019-016-0020-2
• Publication Date: 2016-09-28T03:37:30Z
• Authors: Shahriar Badsha; Xun Yi; Ibrahim Khalil
• Keywords: Homomorphic Encryption; ElGamal encryption; Information sensitivity

The main goal of a personalized recommender system is to provide useful recommendations on various items the users. In order generate recommendations, service needs access types user data such as previous product purchasing history, demographic and biographical information. However, users are sensitive disclosure personal information it can be easily misused by malicious third parties. Consequently, there unavoidable security concerns which will become known through attempted unauthorized while providing recommendation services. protect against breaches information, necessary obfuscate means an efficient encryption technique simultaneously generating making true inaccessible system. To address these challenges, we propose privacy-preserving using homomorphic encryption, without knowing actual ratings. Our approach based ElGamal cryptosystem both addition multiplication plaintexts performed. performance proposed scheme shows significantly high accuracy in-terms computation communication costs well outperforming other existing solutions.