期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：15
页码：3422
出版社：Journal of Theoretical and Applied
摘要：Privacy preserving knowledge discovery is a new and very important topic in data mining that is perfectly talked about the privacy of data and information. This paper focuses on protecting the knowledge in the clustering data mining techniques (K-Mean Clustering). Moreover, a new algorithm is suggested for protecting sensitive clusters, which uses the Adaptive noise techniques for protecting process. In the proposed algorithm, the adaptive noise values that are used for protecting sensitive clusters are evaluated depending on the original database values. In deep, the evaluated noise values depend on the distances (Euclidian Distance) between Sensitive Cluster and the rest of the other clusters (Non-Sensitive Clusters) for the original database. The proposed algorithm use three different techniques for protecting sensitive cluster. The prototype system was used to perform the proposed algorithm. For the three different datasets that are used in a prototype system implementation, the experimental results show that the proposed algorithm is protecting Sensitive Clusters with High Privacy Ratio and Low Information Loss Ratio. Hence, the proposed system provides a good accuracy with a low ratio of side effects, and it supports high level of privacy.
