Abstract

Nowadays, the development of big data, cloud computing, and the internet of things has led to an increase in sharing data. Through the data mining process, some valuable information can be discovered from such shared data. However, most shared data contain personal sensitive information such as users’ location information or disease status and attackers, by analysing such data, may also extract some private (sensitive) information of the user and this can result in threats against the user's privacy. Therefore, before sharing data or making it open we must apply privacy techniques to protect the sensitive information in the data. In this paper, we propose a new approach as well as a technique to guarantee k-anonymity, the most popular privacy protection technique, in the data. The main idea is to design an algorithm to organize tuples/records in the data into groups and then migrate tuples between the groups such that all the groups satisfy k-anonymity. Specifically, the proposed algorithm also maintains the significant association rules in the k-anonymity data so that the data mining process, based on association rule mining, can preserve valuable information as in original data. We perform experiments to evaluate the performance and data utility of our proposed technique in comparison with state-of-the-art anonymization techniques. The experimental results show that our technique outperforms such state-of-the-art ones.