Recommender systems select the most interesting products for costumers based on their interests. The move of a recommender system to a cloud faces many challenges from the perspective of the protection of the participants. Little work has been done regarding secure recommender systems or how to cope with the legal liability of the cloud provider and any virtual machine inside the cloud. We propose DocCloud, a recommender system that focused on the protection of all participants against legal attacks. We present the architecture of DocCloud and analyze the security mechanisms that the system includes. Specifically, we study the properties of plausible deniability and anonymity of the recommenders and intermediate nodes. This way, nodes can recommend products to the customers while deny any knowledge about the product they are recommending or their participation in the recommendation process.

Peer Reviewed