摘要
针对传统的网络流量预测模型精度较低和泛化能力较弱等问题,提出了一种基于K-means聚类算法和支持向量机SVM的回归预测模型KM-SVM。该模型首先利用K-means聚类算法对数据集进行聚类,然后利用SVM训练回归模型,实现对未来流量数据的预测。利用某电商平台的数据中心网络流量数据集进行仿真。实验结果表明,在平均绝对百分比误差和拟合优度等预测精度评价指标上,KM-SVM模型均高于基于BRICH和SVM的预测模型,提高了预测精度。
To solve the problems of low accuracy and weak generalization ability of the traditional network traffic prediction models,a regression prediction model of KM-SVM was proposed based on the K-means clustering algorithm and support vector machine SVM.This model firstly used the K-means clustering algorithm to cluster data sets,and then used SVM training regression model to predict the future traffic data. The network traffic data set of an e-commerce platform was used for simulation. The experimental results showed that,the prediction evaluation indicators such as average absolute percentage error and fit goodness of the KM-SVM model was better than those of the prediction model based on BRICH and SVM,improving the prediction accuracy.
作者
李炜东
李宏慧
LI Weidong;LI Honghui(College of Computer and Information Engineering,Inner Mongolia Agricultural University,Hohhot 010011,China)
出处
《内蒙古农业大学学报（自然科学版）》
CAS
2022年第2期93-97,共5页
Journal of Inner Mongolia Agricultural University(Natural Science Edition)
基金
国家自然科学基金项目(61363016)
教育部留学回国人员科研启动基金项目([2014]1685)
内蒙古自治区自然科学基金项目(2015MS0605,2015MS0626,2020MS06011)。
