Generating Privacy-Preserved Recommendation Using Homomorphic Authenticated Encryption
Piyush Kumar Shanu, K. Chandrasekaran · 2016
Online service providers started to providepersonalized recommendation to the users by collecting userprivate sensitive data. Traditionally the user private data is encrypted using a symmetric encryption algorithm before storing it in the cloud to provide another layer of security for data at rest. It makes users' data secure from third parties, but not the service provider. We propose a method that generates recommendations using homomorphically encrypted data in a privacy preserved manner to provide protection against service provider. We also verify the correctness of computations doneby a third parties and the service provider over encrypted data using homomorphic authenticators and some secure cryptographic protocols.