Improvement of Density Clustering Based k-Anonymity Method
Clustering techniques have recently been successfully adapted for k-anonymity.The anonymous dataset must preserve as much information as possible.To minimize the information loss due to anonymity,it is crucial to group similar data together and then to anonymize each cluster individually.In this paper,an improved density clustering based approach for k-anonymity was proposed.This approach consists of two steps.First,construct clusters of data set using density clustering with cluster size k.Then,adjust each cluster size,let them varying between k and 2k — 1 on condition that all clusters information loss are the smallest.Experimental results show that this approach can further reduce the information loss than previous proposed clustering algorithms.
data publishing k-anonymity density clustering privacy preserving
Xinjun Qi Mingkui Zong
School of software Harbin University Harbin,150086,China
国际会议
秦皇岛
英文
322-325
2010-11-05（万方平台首次上网日期，不代表论文的发表时间）