【摘要】 针对传统监控方法中检测故障的时间比较长的问题,设计了基于网络数据分析的物联网终端监控。利用聚类算法实现物联网终端数据的聚类,计算样本数据的距离,以此得到离群因子的阈值,从而有效处理物联网终端数据,以此能够得出物联网终端中出现问题的数据。实现异常数据的监测、分类和识别,实现将网络数据作为基础的物联网终端监控设计。最后,对系统和传统方法进行对比,对本文设计监控的有效性验证,表示能够使物联网终端故障的检测时间得到缩短,使用价值较高。更多还原

【Abstract】 Aiming at the problem of long time of fault detection in traditional monitoring methods,the terminal monitoring of Internet of things based on network data analysis is designed. The clustering algorithm is used to cluster the terminal data of the Internet of things,calculate the distance of the sample data,and get the threshold value of the outlier factor,so as to effectively process the terminal data of the Internet of things,so as to get the problem data in the terminal of the Internet of things.Realize the monitoring,classification and identification of abnormal data,and realize the design of Internet of things terminal monitoring based on network data. Finally,the system is compared with the traditional methods,and the effectiveness of the design monitoring is verified,which means that the detection time of the Internet of things terminal fault can be shortened,and the use value is higher.更多还原