Granulation as a Privacy Protection Mechanism

How to achieve a balance between data publication and privacy protection has been an important issue in information security for several years. When microdata is released to users, attributes that clearly identify individuals are usually removed. Nevertheless, it is still possible to link released data with some public or easy-to-access databases to obtain confidential information. To safeguard privacy, numerous techniques, such as generalization, suppression, and microaggregation, have been proposed to modify the to-be-released data. In this paper, we propose attribute-oriented granulation as a data protection mechanism that can integrate both generalization and microaggregation into a uniform framework. We address the computational issue of searching for the most specific granulation that satisfies confidentiality requirements. A breadth-first search algorithm with basic pruning strategies is presented and its properties are investigated. The properties can be used to improve the efficiency of our algorithm. We also define some quantitative measures of data quality and security, and apply evolutionary computation techniques to find the optimal granulation for privacy protection.