面向电力工控网络大数据的微聚集差分隐私保护方法

Micro-aggregation for differential privacy protection method based on big data of power control network

针对隐私泄露问题,该文提出一种在频繁模式挖掘中依托微聚集算法实现的差分隐私保护方法,并将其应用到电力工控网络中。通过对指数机制和每个模式的微聚集权重的权衡,选择了Top-k频繁模式方法,并加入拉普拉斯噪声进行扰动,使每个被选择模式的原始支持度均实现了隐私保护与效用的平衡,最大程度地确保了信息发布、数据分析需求和隐私保护需求的平衡,保障了各方对电力工控系统的信任和电力工控系统的健康成长,在数据集上的实验结果验证了该方法的有效性。

In order to solve the problem of privacy disclosure of network,the differential privacy protection in frequent pattern mining is implemented based on the micro-aggregation algorithm for the power industrial control network to ensure the balance among information release,data analysis requirements and privacy protection demands by weighting the exponential mechanism and the micro-aggregation weight of each mode.By adding the Laplace noise disturbance,the Top-k frequent pattern method is selected and the original support each of the selected mode achieves a balance between privacy and utility.The method in this paper can guarantee the trust of all parties in the power industrial control system and the healthy growth of power industrial control system.The experimental results on the dataset verify the effectiveness of the method.