摘要
针对传统隐私保护模型对个性化匿名缺乏考虑的问题,对现有的两种个性化匿名机制进行了分析。在k-匿名和l-多样性匿名模型的基础上,提出一种个性化(α,l)-多样性k-匿名模型来解决存在的问题。在该模型中,依据敏感程度的不同,对敏感属性的取值划分类别;设置相应的约束条件,并为特定的个体提供个性化的隐私保护。实验结果表明,所提模型在有效提供个性化服务的同时,具有更强的隐私保护能力。
Aiming at the problem that traditional privacy preservation model is lack of considering the personalized anonymity,this paper analyzed the existing two personalized anonymity mechanisms.On the basis of k-anonymity and l-diversity model,a personalized(α,l)-diversity k-anonymity model was proposed to solve the existing problems.In the proposed model,the sensitive attribute values are divided into several categories according to their sensitivities,each cate-gory is assigned with corresponding constraints,and the personalized privacy preservation is provided for specific individuals.The experimental results show that the proposed model can provide stronger privacy preservation while supp-lying personalized service efficiently.
作者
曹敏姿
张琳琳
毕雪华
赵楷
CAO Min-zi;ZHANG Lin-lin;BI Xue-hua;ZHAO Kai(College of Information Science and Engineering,Xinjiang University,Urumqi 830046,China;Department of Medical Engineering and Technology,Xinjiang Medical University,Urumqi 830011,China)
出处
《计算机科学》
CSCD
北大核心
2018年第11期180-186,共7页
Computer Science
基金
国家自然科学基金(61562088)
新疆维吾尔自治区科技厅项目(2017D01C232)
新疆维吾尔自治区高校科研计划项目创新团队(XJEDU2017T002)
新疆维吾尔自治区高校计划项目(XJEDU2017M005)
赛尔网络下一代互联网技术创新项目(NGII20170325)资助
关键词
隐私保护
K-匿名
l-多样性
个性化匿名
泛化
Privacy preservation
k-anonymity
l-diversity
Personalized anonymity
Generalization