嵌入式异构物联网敏感数据流动态挖掘研究

Research on dynamic mining of sensitive data flow in embedded heterogeneous IoT

数据召回速率影响网络主机对敏感数据流的挖掘与处理能力。因此,为提升数据召回速率,针对嵌入式异构物联网敏感数据流动态挖掘方法展开研究。以边缘云架构为基础,推导数据标识映射条件,并完善嵌入式异构物联网体系。然后根据滑动窗口构建标准,建立闭合频繁项集合,再通过求解负载配比参数的方式定义关联挖掘规则,从而实现对敏感数据流的动态挖掘。实验结果表明,该方法能够将数据召回速率最大值提升至8.0 bit/ms,使得物联网主机对敏感数据流的挖掘与处理能力得到了提升。

The data recall rate affects the mining and processing ability of network hosts to sensitive data streams.Therefore,in order to improve the data recall rate,dynamic mining methods for sensitive data flow of embedded heterogeneous IoT are studied.Based on the edge cloud architecture,the data identification mapping conditions are derived,and the embedded heterogeneous Internet of Things system is improved.Then,according to the sliding window construction standard,the closed frequent item set is established,and the association mining rules are defined by solving the load matching parameters,so as to realize the dynamic mining of sensitive data streams.The experimental results show that this method can increase the maximum data recall rate to 8.0 bit/ms,which improves the ability of the host of the Internet of Things to mine and process sensitive data streams.