Research On Semantic Expansion Based Fuzzy Search Over Encrypted Data In Cloud

With the rapid development of cloud computing, more and more companies and individuals choose to store their data in the cloud. Cloud storage reduces the burden of storage management on the user’s side, and allows the users to access data anywhere. It provides users with flexible storage services. With cloud storage model, the user’s data needs to be outsourced to cloud server for storage and management. However, cloud is a third-party platform, and can easily access the data stored by the user. Thus user’s privacy and sensitive data may be leaked. Security has been hindered the development of cloud computing.To protect data privacy and prevent unauthorized access, the sensitive data has to be encrypted by data owner before being outsourced. When the authorized users want to retrieve certain data files, they have to download all the encrypted data files, and decrypt them to obtain what they want. However, considering the large scale of data files stored in the cloud, it is not a efficient way to retrieve the specific files. If the encrypted data files couldn’t be searched and utilized easily, it violate the goal of data sharing in cloud computing. Efficient search over encrypted data has been an urgent problem need to be addressed. Therefore, the study of search over encrypted cloud data is important.The existing techniques only support exact search, namely the search keyword should exactly match the words in the index. Some documents which contain semantically related words are ignored. For the reason that users have limited domain knowledge, the query keywords they could provide are limited, which can’t fully express the actual query intent of users. Thus the search results are incomplete and inaccurate. For the deficiencies, we propose a fuzzy search solution over encrypted cloud data based on semantic expansion. The main contributions are described below.1) We introduce plaintext semantic query expansion technology into the domain of ciphertext retrieval. In the retrieval process, the submitted query keyword is semantically expanded to get the query keywords set. Then the keywords are used for retrieving to improve the recall. The search results are more suitable to the users’ query intend. Based on the idea above, a new semantic expansion search scheme in the domain of ciphertext retrieval is proposed. The scheme could restrict the existing threat model. In the solution, the encrypted file metadata set is uploaded to the cloud server. We construct the inverted index with the encrypted metadata set to improve the efficiency. The semantic relationship library is also constructed to implement the semantic query expansion. In addition, we modify the order-preserving encryption algorithm to ensure the security and preserve the rank capability. In the end, we implement the semantic expansion based search system, and evaluate the performance with the real dataset.2) We find a new attack-the threat of co-occurrence relationship between terms, which proves that the existing schemes using the inverted index leak privacy information. In view of the security problem, we propose a more secure solution based on hybrid cloud model. In the scheme, little information storage and pre-encryption are done by the private cloud, and large-scale ciphertext storage and query operation with large computation overhead are completed by the public cloud. The solution improves the security and reduces the service cost at the same time, which improves the retrieval performance. We construct the file vector and query vector based on the vector space model. The relevance sore between the file and query is evaluated by calculating the similarity of vectors. Then we can rank the result documents with the relevance score. Eventually, we implement the more secure semantic expansion based search system, and evaluate the performance with the real dataset.