Abstract

Nowadays, data mining techniques play a major role in obtaining useful data from large amounts of data. Released information may include personal or sensitive information. It is necessary to protect this sensitive information from unauthorized access. This is why, privacy protection becomes an important part of data mining. Over the years, various privacy protection strategies have evolved to protect personal information. After all, Anonymization is one of the most important ways to maintain privacy. A variety of common anonymous methods are used to maintain privacy, however these methods have some flaws in maintaining personal privacy. The Framework for Efficient Anonymous Algorithm is therefore proposed here. Initially, the proposed algorithm aims to identify critical and sensitive data characteristics using the Principal Component Analysis based Attribute Selection Algorithm. The algorithm estimates Eigen values and Eigen vectors. Over time, the process of encryption was performed by introducing the Novel Based Anonymity Algorithm (NBA). Finally anonymous information is available that prevents unauthorized access to personal information. To evaluate the performance of the proposed algorithm many factors such as data usage, privacy levels and computer costs are compared to existing systems. From the experimental analysis, the effectiveness of the proposed system proves its superiority compared to other strategies.

Key words: Data Mining, Privacy Preservation, K-Anonymity.