基于K-means聚类的电力大数据脱敏技术研究

Research on desensitization technology of power big data based on K-means clustering

针对传统大数据脱敏技术缺少对动态数据的脱敏处理,导致大数据脱敏效果不佳的问题,提出基于K-means聚类的电力大数据脱敏技术研究。根据电力数据异常量的K-means聚类中心距离进行检测,建立脱敏系统,在系统中预留元数据管理接口,依据检测结果导入敏感配置信息。根据用户需求,通过对敏感数据的识别、流程的判断,确定敏感数据,对静态和动态敏感数据进行脱敏处理。实验结果可知,该技术只暴露用户身份证号前两位,其余信息均能被脱敏处理,有效保证了用户身份信息安全。

Aiming at the problem that the traditional big data desensitization technology is lack of dynamic data desensitization processing,which leads to the bad effect of big data desensitization,this paper proposes the research of power big data desensitization technology based on K-means clustering. According to the K-means clustering center distance of power data anomaly detection,the desensitization system is established,the metadata management interface is reserved in the system,and the sensitive configuration information is imported according to the detection results. According to the needs of users,through the identification and judgment process of sensitive data,the sensitive data is determined,and the static and dynamic sensitive data are desensitized. The experimental results show that the technology only exposes the top two users’ ID number,and the rest of the information can be desensitized,effectively guaranteeing the user identity information security.