Privacy Preserving Data Mining Technique to Secure Distributed Client Data

Abstract

AbstractData is now acquired and stored in databases from a variety of sources. The collecting of data has little influence until the database owner does data analysis, such as by applying data mining techniques to the databases. Data mining techniques and algorithms are currently being developed, and they are providing substantial improvements to the information extraction process. With respect to data mining, privacy-preserving data categorization is an important study area. Data mining-based privacy attacks, which entail analyzing data over a lengthy period of time to retrieve useful information, are one of the cloud’s security issues. In this paper, we introduce concept of privacy preserving data mining. We investigate the privacy and security issues and propose a privacy preservation data mining approach. The Proposed Privacy Model for distributed client has been proposed using decision tree with resulting parameter accuracy, error rate, time complexity and space complexity. At last to validate the performance of the proposed approach, we compare by traditional PPARM approach is used with similar dataset for comparison. The experimental outcome shows that the effectiveness of the proposed algorithm of privacy preserving that secure multi-party computation can be practically, even for compound problems and large size inputs.KeywordsPrivacy preservingC 4.5Distributed databaseSensitive informationMultiparty computationData releasePPDM