数据发布匿名技术进展  

Advancement of anonymity technique for data publishing

作　　者：刘湘雯[1] 王良民[1] 

机构地区：[1]江苏大学计算机科学与通信工程学院,江苏镇江212013

出　　处：《江苏大学学报（自然科学版）》2016年第5期562-571,共10页Journal of Jiangsu University：Natural Science Edition

基　　金：国家自然科学基金资助项目(61472001);江苏省重点研发计划项目(BE2015136);镇江市工业支撑计划项目(GY2013030)

摘　　要：随着计算机技术和网络技术的快速发展,以数据挖掘与分析为目的的数据发布实现了信息的高度共享,但也因此造成数据中包含的大量敏感隐私信息的泄漏风险.匿名技术是解决数据发布中的隐私泄露问题的主要方法.首先简要介绍数据发布隐私保护中的数据匿名化处理场景;其次分别对处理单敏感、多敏感属性的静态数据发布、增量数据发布、数据流发布、轨迹数据发布的匿名模型进行归纳,总结匿名模型对不同的攻击形式如链接攻击、同质攻击、背景知识攻击等的抵御能力;接着分析比较了泛化、抑制、聚类、微聚集、分解、置换等匿名化方法;然后讨论了匿名技术在数据聚合、位置服务、社交网络等领域的发展;最后总结并指明下一步的研究方向.With the rapid development of computer and network technology, data publishing achieves well-sharing of information for the purpose of data mining and analyzing, whille it may cause the leakage risk of a great amount of sensitive privacy information in the data. Anonymity technology is the main method to realize privacy preservation in data publishing. The research progress of anonymity technology in data publishing was summed up. The scenario of data anonymization was introduced briefly. The anonymization models were summarized, which were respectively used to deal with static data publishing of single sensitive attribute and multidimensional ones, incremental data publishing, streaming data publishing and trajectory data publishing. The anonymization model abilities to resist linking attack, homogeneity attack and background knowledge attack were analyzed. The anonymization methods of generalization, suppression, clustering, microaggregation, anatomy and permutation were analyzed and compared. The applications of anonymity techniques for privacy preservation in the fields of data mashup, location based services and social network were discussed. The future research of anonymity technology for data publishing was proposed.

关 键 词：数据发布 隐私保护 匿名化 K-匿名 轨迹数据 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]