ENHANCE DATA SECURITY IN CLOUD COMPUTING USING MACHINE LEARNING AND HYBRID CRYPTOGRAPHY TECHNIQUES

Abstract

In Cloud computing, the user’s can access their important data via internet that is being stored on the remote servers. As technology is growing day by day, there is a rapid increase in the personal and crucial data. This brings up the need of securing the users data. Data can be of any type and each required different degree of protection. In this paper, we have proposed a model that classifies the data according to its security parameters. The performance of the existing KNN is improved by appending it with ensemble learning technique. The basic algorithms of ensemble learning i.e., base level-0 and meta level-1 are modified. This will improve prediction capability and classification accuracy of existing KNN technique. Also to secure sensitive data, the hybrid cryptography technique is used. The quantitative analysis show that the accuracy of ensemble learning when combines with existing KNN is 73.5% whereas the accuracy of KNN was 65.5%.