PROJECT TITLE :

Privacy Preserving Multi-keyword Top-k Similarity Search Over Encrypted Data - 2017

ABSTRACT:

Cloud Computing provides people and enterprises large computing power and scalable storage capacities to support a variety of massive information applications in domains like health care and scientific analysis, thus a lot of and a lot of knowledge house owners are concerned to outsource their information on cloud servers for nice convenience in information management and mining. However, knowledge sets like health records in electronic documents typically contain sensitive data, that brings regarding privacy issues if the documents are released or shared to partially untrusted third-parties in cloud. A sensible and widely used technique for knowledge privacy preservation is to encrypt information before outsourcing to the cloud servers, which however reduces the data utility and makes several ancient information analytic operators like keyword-based high-k document retrieval obsolete. In this paper, we tend to investigate the multi-keyword top-k search drawback for giant knowledge encryption against privacy breaches, and attempt to spot an efficient and secure answer to the present problem. Specifically, for the privacy concern of query information, we tend to construct a special tree-primarily based index structure and style a random traversal algorithm, which makes even the same query to provide totally different visiting paths on the index, and will conjointly maintain the accuracy of queries unchanged beneath stronger privacy. For improving the query efficiency, we propose a cluster multi-keyword high-k search theme based mostly on the thought of partition, where a cluster of tree-based mostly indexes are created for all documents. Finally, we tend to mix these strategies along into an efficient and secure approach to deal with our proposed prime-k similarity search. In depth experimental results on real-life information sets demonstrate that our proposed approach can significantly improve the aptitude of defending the privacy breaches, the scalability and also the time potency of query processing over the state-of-the-art strategies.