您的位置：首页 > 论文页面
多维敏感属性流数据发布的隐私保护方法
发表时间：2021-06-19 浏览量：750 下载量：146
| 全部作者： | 程林，丰江帆 |
| 作者单位： | 重庆邮电大学计算机科学与技术学院 |
| 摘 要： | 在很多应用场景下，数据是实时发布的，并且这些数据绝大部分由多个敏感属性组成，针对传统的多维敏感属性数据发布隐私保护方法主要解决静态多维敏感属性数据发布或数据动态连续发布的隐私保护，而对多敏感属性实时流数据发布的研究较少这一问题，本文提出一种基于滑动窗口模型的加权优化多维桶分组算法（weighted op-timization multi-dimensional bucket grouping algorithm，WOMBPA）。首先，在实时流数据发布过程中基于滑动窗口模型批处理思想，对不同敏感属性进行排序；然后根据不同属性敏感程度值，对数据构建加权优化多维桶分组；最后进行匿名化处理。实验分析结果表明，本文所提方法解决流数据发布中多个敏感属性保护的同时，也减少了信息损失度和隐匿率，保证发布数据的可用性。 |
| 关 键 词： | 计算机软件；隐私保护；多敏感属性；滑动窗口；加权优化多维桶 |
| Title： | Privacy protection method for multi-dimensional sensitive attribute streaming data publication |
| Author： | CHENG Lin, FENG Jiangfan |
| Organization： | College of Computer Science and Technology, Chongqing University of Posts and Telecommunications |
| Abstract： | In many application scenarios, data is published in real time, and most of these data are composed of multiple sensitive attributes. The privacy protection method for traditional multi-dimensional sensitive attribute data publication mainly solves the privacy protection of static multi-dimensional sensitive attribute data publication or dynamic continuous data publication. However, there are few researches on the real-time streaming data publication of multi-sensitive attributes. To solve this problem, a weighted optimization multi-dimensional bucket grouping algorithm (WOMBPA) based on sliding window model is proposed. Firstly, the different sensitive attributes are sorted in the real-time streaming data publishing process based on the batch processing idea of the sliding window model. Then, according to the sensitivity values of different attributes, the weighted optimization multi-dimensional bucket grouping is constructed for the data. Finally, anonymization is performed. Experimental analysis results prove that the method proposed in this paper solves the protection of multiple sensitive attributes in streaming data publication, while also reducing the degree of information loss and supression ratio, and ensuring the availability of published data. |
| Key words： | computer software; privacy protection; multiple sensitive attributes; sliding window; weighted optimization multi-dimensional bucket |
| 发表期数： | 2021年6月第2期 |
| 引用格式： | 程林，丰江帆. 多维敏感属性流数据发布的隐私保护方法[J]. 中国科技论文在线精品论文，2021，14（2）：212-219. |
1
评论数 0请您登录
暂无评论
下载全文
多维论文
