KNN Classification of Encrypted Cloud Data with Privacy Preservation

Data Mining has wide applications in different zones, for occurrence, keeping money, medicine, investigative examination and among government work environments. Solicitation is one of the regularly utilized assignments as a bit of data mining applications. As far back as decade, in view of the move of different confirmation issues, different theoretical and sound judgment answers for the solicitation issue have been proposed under arranged security models. Regardless, with the late reputation of passed on figuring, clients now have the chance to outsource their information, in encoded structure, moreover the data mining assignments to the cloud. Taking after the information on the cloud is in encoded structure, existing security guaranteeing depiction frameworks are not suitable. In this paper, we concentrate on comprehension the depiction issue over encoded information. Specifically, we propose a shielded k-NN classifier over mixed information in the cloud. The proposed convention ensures the course of action of information, security of client's data demand, and covers the information access traces. To the best of our taking in, our work is the first to add to a secured k-NN classifier over mixed information under the semi-genuine model. Likewise, we correctly break down the practicality of our proposed convention utilizing a true blue dataset under different parameter settings.