Privacy-preserving Personalized Search over Encrypted Cloud Data Supporting Multi-keyword Ranking

Cloud computing is emerging as a revolutionary computing paradigm which provides a flexible and economic strategy for data management and resource sharing. Security and privacy become major concerns in the cloud scenario, for which Searchable Encryption (SE) technology is proposed to support efficient keyword based queries and retrieval of encrypted data. However, the absence of personalized search is still a typical shortage in existing SE schemes. In this paper, we focus on addressing personalized search over encrypted cloud data and propose a Privacy-preserving Personalized Search over Encrypted Cloud Data Supporting Multi-keyword Ranking(PPSE) scheme that supports Top-k retrieval in stringent privacy requirements. For the first time, we formulate the privacy issue and design goals for personalized search in SE. We introduce the Open Directory Project to construct a formal model for integrating preferential ranking with keyword search reasonably and automatically, which can help eliminate the ambiguity of any two search requests. In PPSE, we employ the vector space model and the secure kNN scheme to guarantee sufficient search accuracy and privacy protection. The tf-idf weight and the preference weight help to ensure that the search result will faithfully respect the user's interest. As a result, thorough security analysis and performance evaluation on experiments performed on the real-world dataset demonstrate that the PPSE scheme indeed accords with our proposed design goals.