Date of Publication :7th September 2015

Abstract: Cloud computing is recent commonly used method for data outsourcing through web services. We propose a concept to search any encrypted document in cloud store based on multiple keywords specified with the document. Keywords for the document can be of contents from the document, which are selected while uploading that file to the cloud server. The result retrieved after searching is a list of k files (index file). File selection is based on the score calculated by using TRSE (Two Round Searchable Encryption) algorithm. The TRSE is a combination of vector space model and homomorphic encryption technique. Since the homomorphic encryption allows users to do majority of their work on server side by operation only on cipher text. We can eliminate information leakage and can ensure data security. In this paper we propose a concept to score these documents more efficiently on the basis of search history of the document with TRSE scheme. The search history may include the location where the document is being accessed or the number of access to the document by the same user. Besides this when the owner try to update document, with the help of hashing we notify the owner to update index file to avoid data inconsistency and ensure efficiency.

Reference :