The Research Of Privacy-Preserving Multi-Keyword Search Schema Over Encrypted Cloud Data

With the development of cloud computing,services outsourcing in clouds has become a popular business model.However,duo to the fact that data storage and computing are completely outsourced to the cloud service provider,sensitive data of data owner is completely exposed,which could bring serious security problem of privacy disclosure.Data owner often encrypts their own document data and then outsource it to cloud service provider.However,the category relationship has been concealed by the traditional searchable encryption method,resulting in unnecessary computational overhead.Hence,these methods do not adapt to the cloud computing scenario where data storage is very large and online data processing is the basic requirement.Therefore,it is desirable to utilize the category relationship to perform efficient and accurate search for encrypted outsourced documents.In this thesis,we focus on the problems of privacy-preserving multi-keyword search technology over encrypted cloud data.From the two aspects of efficient retrieval and accurate retrieval,we propose an efficient and secure multi-keyword search technology(ESMS)and an accurate and secure multi-keyword search technology(ASMS).CCV-tree,the index structure of ESMS,is constructed based on the sequence of content-related documents.The documents with high relevance are gathered in the same sub-tree.The root of sub-tree stores the cluster center vectors of the corresponding clusters.Meanwhile,binary search algorithm is proposed,which only selects more relevant sub-clusters at each time to achieve efficient multi-keyword search.ASMS adopts the idea of hierarchical agglomerative clustering to construct a novel EMV-tree index,which has higher document relevance in the same sub-tree.Based on EMV-tree,a non-candidate prune depth first search algorithm(NCPDFS)is presented.NCP-DFS can quickly filter out subtrees that can not exist Top-k documents.ASMS greatly accelerates the search efficiency without losing accuracy.Secure KNN technology is applied to ESMS and ASMS to encrypt index vectors and search vectors to realize secure multikeyword ciphertext retrieval.The theoretical analysis and experimental results show that ESMS can realize the retrieval time grows linearly with the exponential growth of the number of documents,and achieve efficient and secure multi-keyword ciphertext retrieval.ASMS outperforms existing methods in terms of efficiency without losing accuracy and achieve more efficient and secure multi-keyword ciphertext retrieval.