An Efficient Privacy Preserving using Map Reduce based International Data Encryption Algorithm and weighted Auto Encoder

Abstract

One of the most important aspects of big data investigation is data security. The majority of cloud system applications involve sensitive data, namely personal, business, or health records. Threats to such data could put the cloud platforms that store it in jeopardy. Conventional security solutions, on the other hand, are incapable of securing big data migration. To handle the generation of vast amount of data and security aspects of generated data across cloud is handled by an effective privacy preserving mechanism. Initially, cloud based dataset is clustered and information are balanced with map-reduce mechanism. Further, the information are encrypted using International Data Encryption Algorithm (IDEA) and the convolution process is attained over pertain the estimation process to the encrypted or convoluted data. The estimation process is accomplished by the weighted Auto encoder KNN (WAEKNN) classifier. Suppose the encrypted data are not appropriately encrypted means the data are again transmitted to the convolution process. Experimental results of proposed framework is compared with existing technique and the big data based privacy preserving scheme outperforms the existing techniques.