An Efficient Multi-Keyword Searching with User Confidentiality in Information Networks

Abstract

In Information Networks, owners can store their documents over distributed multiple servers. It facilitating users to store and access their data in and from multiple servers by sitting anywhere and on any device. It is a very challenging task to provide efficient search on distributed documents also provide the privacy on owner’s documents. The existing system provides one possible solution that is privacy preserving indexing (PPI). In this system, documents are distributed over multiple private servers which are collectively controlled by cloud/public server. When user wants some documents, they query to public cloud, which then returns the candidate list that is private server list to users. After getting list, user can search the documents on specific private server but in this system, documents are stored in plain text form on private server that is privacy is compromised. But proposed system enhances this existing system to make it more secure and efficient. First documents are stored in encrypted form on the private servers and then use Key Distribution Center (KDC) for allowing decryption of data receive from private server, at client side. The proposed system also implements TF-IDF, which provides the ranking of results to users.