Enhancement in Classification of Semantically Secure Encrypted Data

Abstract

-Techniques which are already existing are not having the high clustering to the encrypted data on cloud, therefore, to solve the classification problem on encrypted data. A k-NN algorithm for classification of encrypted data in the cloud, is also proposed for security purpose. In this paper we proposed the the new DBSCAN algorithm which will provide the better clustering than the K-NN clustering of the data on the cloud, provides In day by day the popularity of web services are attracting with their rapid development, As a result, there is a huge amount of heterogeneous data. The data needs to be mine for various applications in many organizations like scientific research, medicine and among government agencies. Data Mining is perspective is on a very large scale. Classification of outsourced data is One of the most important tasks in data mining applications. So in data mining area many practical as well as theoretical solutions to the classification problem have been proposed. As as a privacy issues solution, Different security models used in different solutions. The recent needs of IT make Cloud Computing to exist. Always users can outsource encrypted form of their data in and the data mining tasks to the cloud. Privacy preserving classification security to user's input query and hiding the data accessing patterns on the cloud. As well as it will provide better clustering solution for the outsourced data.