A Verifiable and Dynamic Multi-keyword Ranked Search Scheme over Encrypted Cloud Data with Accuracy Improvement

With the widely application of cloud computing, more and more data owners prefer to outsource their data on the remote cloud servers to reduce the overhead. Searchable encryption is proposed in an urgently need for searching on the encrypted data. In this paper, we present a tree-based privacy-preserving and efficient multi-keyword ranked search scheme supporting verification and dynamic update. Considering the effect of the keywords location on the weight in most documents, the traditional \(TF\times IDF\) algorithm can be optimized with location information to get more accurate similarity score. To improve the efficiency, we combine the vector space model and binary tree to construct a tree-based index structure. And the index tree is encrypted by secure kNN computation. Finally, We analyze the security against two threat model, and implement the experiment on the real paper set to evaluate the performance.