基于集成学习的物联网通信数据快速分类研究  

Research on Rapid Classification of Internet of Things Communication Data Based on Ensemble Learning

作　　者：杨瑞丽 王俊仃 邱秀荣[1] YANG Ruii;WANG Junding;QIU Xiurong(School of Information and Electronic Engineering,Shangqiu Institute of Technology,Shangqiu 476000,China)

机构地区：[1]商丘工学院信息与电子工程学院,河南商丘476000

出　　处：《通信电源技术》2025年第5期4-6,共3页Telecom Power Technology

基　　金：商丘工学院2024年校级科研项目“工业物联网大数据融合技术研究”(2024KYXM05)。

摘　　要：物联网设备持续产出的数据中会掺杂部分异常数据,导致物联网通信数据分类的质量与效率下降。因此,提出一种基于集成学习的物联网通信数据快速分类方法。从物联网设备收集通信数据,利用孤立森林算法确定物联网通信数据样本的异常分值,并去除异常分值较高的数据,通过基于密度的带噪声应用空间聚类(Density-Based Spatial Clustering of Applications with Noise,DBSCAN)算法整合去除异常后的数据,结合集成学习算法实现物联网通信数据快速分类。实验结果表明,所提方法的物联网通信数据分类准确率始终在97.2%以上,物联网通信数据分类时间均值约为1.55 s,具有良好的应用潜力。The data continuously generated by Internet of Things devices may contain some abnormal data,leading to a decrease in the quality and efficiency of Internet of Things communication data classification.Therefore,a fast classification method of Internet of Things communication data based on ensemble learning is proposed.Collecting communication data from Internet of Things devices,using isolated forest algorithm to determine the abnormal scores of Internet of Things communication data samples,and removing the data with higher abnormal scores,integrating the abnormal data by Density-Based Spatial Clustering of Applications with Noise(DBSCAN)algorithm,and combining with ensemble learning algorithm to realize rapid classification of Internet of Things communication data.The experimental results show that the accuracy of the Internet of Things communication data classification by the proposed method is always above 97.2%,and the average time of the Internet of Things communication data classification is about 1.55 s,which has good application potential.

关 键 词：集成学习 物联网通信 数据分类 基于密度的带噪声应用空间聚类(DBSCAN) 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]