PRMS: A Personalized Mobile Search Over Encrypted Outsourced Data

Abstract

At present, searchable encryption is a very promising direction in the field of cloud computing. However, most existing works focus on keyword-based search schemes and regardless of personalized search needs, a keyword-based search does not take into account the user’s location information and cannot exactly match users’ search intentions. Since location information is very important in mobile searches, we propose a personalized mobile search (PRMS) over encrypted outsourced data, and we convert the user’s location information into distance information to generate the user location model, forming a location query matrix with user location information. The matrix is then used to encrypt the user’s location query and conceal the location information in the location query matrix. In this paper, we combine a user’s interest preference and location information in personalized searches over encrypted outsourced data, adopt the law of universal gravitation to calculate scores of files in the cloud, and return the first $K$ results with the highest gravitational forces. For the first time, we propose a PRMS-improvement scheme that is applied to an encryption method to build the content index and location index, which can greatly reduce the time for model construction. Through the PRMS scheme, we realized personalized mobile searches based on user’s interests and location information.