Efficient multi-keyword ranked query over encrypted data in cloud computing

Ruixuan Li, Zhiyong Xu, Wanshang Kang, Kin Choong Yow, Cheng-Zhong Xu
2014 Future generations computer systems  
h i g h l i g h t s • Design a novel storage and encryption algorithm to manage the keyword dictionary. • Greatly reduce both the dictionary reconstruction overhead and the file index re-encryption time as new keywords and files are added. • Design a novel trapdoor generation algorithm. • Take the keyword access frequencies into account when the system generates the ranked list of the returning results. a b s t r a c t Cloud computing infrastructure is a promising new technology and greatly
more » ... lerates the development of large scale data storage, processing and distribution. However, security and privacy become major concerns when data owners outsource their private data onto public cloud servers that are not within their trusted management domains. To avoid information leakage, sensitive data have to be encrypted before uploading onto the cloud servers, which makes it a big challenge to support efficient keywordbased queries and rank the matching results on the encrypted data. Most current works only consider single keyword queries without appropriate ranking schemes. In the current multi-keyword ranked search approach, the keyword dictionary is static and cannot be extended easily when the number of keywords increases. Furthermore, it does not take the user behavior and keyword access frequency into account. For the query matching result which contains a large number of documents, the out-of-order ranking problem may occur. This makes it hard for the data consumer to find the subset that is most likely satisfying its requirements. In this paper, we propose a flexible multi-keyword query scheme, called MKQE to address the aforementioned drawbacks. MKQE greatly reduces the maintenance overhead during the keyword dictionary expansion. It takes keyword weights and user access history into consideration when generating the query result. Therefore, the documents that have higher access frequencies and that match closer to the users' access history get higher rankings in the matching result set. Our experiments show that MKQE presents superior performance over the current solutions. (R. Li). each application can be scaled up and down according to the fluctuating demand. It adopts a pay-per-use resource sharing model, which allows a user to pay only for the number of service units it consumes. Cloud computing infrastructure provides a flexible and economic strategy for data management and resource sharing. It can reduce hardware, software costs and system maintenance overheads. It can also offer a convenient communication channel to share resources across data owners and data consumers. With the popularity of cloud services, such as Amazon Web Services, 1 1 Amazon web services, http://aws.amazon.com.
doi:10.1016/j.future.2013.06.029 fatcat:pdt7jrwobvd6ljft2zx2pfbalm