Micro-Aggregation Algorithm Based on Sensitive Attribute Entropy
Jin Song Yang · Dianzi xuebao · 2014
In the process of clustering,inappropriate distance measure leads to unnecessary loss of information during the anonymous process,so it is a difficult problem to define a proper distance measurement for different types of variables.We put forward the concept of semantic attribute,and propose a coding hierarchy tree to represent semantic attribute and to reduce the information loss in the anonymous process. In the p-sensitive k-anonymity model,the uneven distribution of the sensitive attribute values in the clustering results may cause sensitive information disclosure,so we propose a micro-aggregation algorithm based on sensitive attribute entropy. Moreover,we propose the concept of anonymous protection factor to describe the degree of privacy protection. During the process of clustering,in order to improve the uniformity of the distribution of sensitive attribute values in the clustering results,the algorithm ensures the maximum of anonymous protection factor,so it can deal with the background knowledge attack and reduce the risk of privacy leaking. Finally,the rationality and validity of the algorithm is verified by experiment.