| 基于聚类匿名化的差分隐私保护数据发布方法 |
| |
| 引用本文： | 刘晓迁,李千目.基于聚类匿名化的差分隐私保护数据发布方法[J].通信学报,2016,37(5):125-129. |
| |
| 作者姓名： | 刘晓迁 李千目 |
| |
| 作者单位： | 南京理工大学计算机科学与工程学院，江苏 南京 210094 |
| |
| 基金项目： | 中央高校基本科研业务专项基金资助项目（No.3091605104）；国家自然科学基金资助项目（No.61272419）；江苏省未来网络前瞻性研究基金资助项目（No.BY2013095-3-02）；江苏省产学研前瞻性基金资助项目（No.BY2014089, No.BY2013039, No.BY2013037）；江苏省普通高校研究生创新计划基金资助项目（No.KYLX15_0384） |
| |
| 摘 要： | 基于匿名化技术的理论基础，采用DBSCAN聚类算法对数据记录进行聚类，实现将个体记录匿名化隐藏于一组记录中。为提高隐私保护程度，对匿名化划分的数据添加拉普拉斯噪声，扰动个体数据真实值，以实现差分隐私保护模型的要求。通过聚类，分化查询函数敏感性，提高数据可用性。对算法隐私性进行证明，并实验说明发布数据的可用性。
|
| 关 键 词： | 差分隐私 隐私保护 聚类 数据发布 匿名化 |
Differentially private data release based on clustering anonymization |
| |
| Xiao-qian LIU,Qian-mu LI.Differentially private data release based on clustering anonymization[J].Journal on Communications,2016,37(5):125-129. |
| |
| Authors: | Xiao-qian LIU Qian-mu LI |
| |
| Affiliation: | School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing 210094,China |
| |
| Abstract: | Based on the theory of anonymization, the DBSCAN method was applied to divide all the data records into different groups to cover individuals. To provide privacy enhancement, the Laplace noise was added to the anonymized partitioned data to perturb the real value of data record so that the requirements of differential privacy model were satisfied. With the clustering operation, the sensitivity of the query function has been partitioned to improve data utility. The proof of privacy has been given and experimental results have been provided to evaluate the utility of the released data. |
| |
| Keywords: | differential privacy privacy preservation clustering data release anonymization |
|
| 点击此处可从《通信学报》浏览原始摘要信息 |
|
点击此处可从《通信学报》下载全文 |
