An Efficient Multi-keyword Ranked Retrieval Scheme with Johnson-Lindenstrauss Transform over Encrypted Cloud Data

Cloud computing is becoming more and more popular recently, and this trend encourages large corporations and companies to outsource their data to a remote cloud server. However, the concern of data privacy and security is a huge problem which impedes the development of cloud computing. It is essential and necessary to propose a scheme to realize the information retrieval and protect the data privacy at the same time. Since ranked multi-keyword search is a pretty practical question in this area and it has a significant influence on the user experience, we would mainly focus on the scheme to achieve high performance and low cost multi-keyword ranked search. In this paper, we proposed a novel search scheme with Johnson-Lindenstrauss transform (J-L scheme for short) and a technique called "optimised maximum query" to live up to our efficiency demand. Furthermore, we established a complete system to guarantee a set of strict privacy requirements. Extensive experiments on real-world database are presented to show our proposed scheme\´s efficiency and accuracy.