PROJECT TITLE:

k-Nearest Neighbor Classification over Semantically Secure Encrypted Relational Data - 2015

ABSTRACT:

Knowledge Mining has wide applications in many areas like Banking, medication, scientific analysis and among government agencies. Classification is one in all the commonly used tasks in information mining applications. For the past decade, thanks to the increase of various privacy problems, many theoretical and practical solutions to the classification drawback are proposed beneath totally different security models. However, with the recent popularity of Cloud Computing, users currently have the opportunity to outsource their knowledge, in encrypted type, along with the info mining tasks to the cloud. Since the information on the cloud is in encrypted type, existing privacy-preserving classification techniques aren't applicable. In this paper, we concentrate on solving the classification downside over encrypted knowledge. In specific, we propose a secure k-NN classifier over encrypted data in the cloud. The proposed protocol protects the confidentiality of data, privacy of user's input query, and hides the information access patterns. To the most effective of our information, our work is the first to develop a secure k-NN classifier over encrypted data underneath the semi-honest model. Conjointly, we tend to empirically analyze the efficiency of our proposed protocol employing a real-world dataset beneath totally different parameter settings.