A Novel Anonymity Model Based on Information Entropy and Hierarchical Sequential Three-Way Decisions
Mingchen Zheng, Jin Qian, Chuanpeng Zhou, Di Wang, Yuehua Lu, Changlang Shuai · 2024
In the era of big data, addressing privacy risks has become a critical concern, this paper proposes a new anonymity model based on information entropy and hierarchical sequential three-way decisions. Firstly, we use hierarchical sequential three-way decision to partition the original dataset into multiple granularity levels based on quasi-identifier attributes. Then we calculate the information entropy of each granularity level to determine its priority. Subsequently, we process the granularity spaces sequentially, classifying the data into disjoint parts within a single granularity space using the three-way decisions and adopting the different anonymization strategies. Finally, We performed a comparison of the proposed algorithm with two other algorithms on six datasets. The results of experiments demonstrate that our algorithm enhances anonymization efficiency while ensuring data privacy, and proves feasible and effective in terms of runtime and processing cost.