基于大数据模式分解的隐私信息保护方法仿真

Research on-Line Monitoring Method of Noise Pollution in Robot Master-Slave Control system

针对传统的大数据隐私保护算法执行效率慢、适应性差的问题,提出一种基于大数据模式分解的隐私信息保护方法,首先通过模式分解、安全分解以及正确分解,建立大数据分解DMM矩阵,生成一个随机满足的网络隐私变换函数,转换初始网络隐私数据,并将其结果视为随机回答结果,利用属性聚类的等价划分方式,聚类准标识符属性类似值与敏感属性类似值,两者属性通过实现等价类划分与数据的匿名操作,完成隐私信息保护。仿真证明,所提方法相比传统方法的保护效果更好,执行率更高。

Due to the slow execution rate and poor adaptability of traditional big data privacy protection algorithm, this paper presented a method for protecting privacy information based on big data pattern decomposition. Firstly, the big data decomposition DMM matrix was built by pattern decomposition, security decomposition, and correct decomposition. And then, a random network privacy transformation function was generated to convert the initial network privacy data and treat the result as a random answer. Moreover, the equivalent partitioning method based on attribute clustering was used to integrate with the similar values of the quasi-identifier attribute and sensitive attribute. Finally, the protection of private information was completed by implementing the equivalent partition and the anonymous operation of data. Simulation results prove that the proposed method has a better protection effect and higher execution rate than traditional methods.