机器学习在网络优化中的应用研究  

Application of Machine Learning in Network Optimization

作　　者：王雨 刘琳琳[1] WANG Yu;LIU Lin-lin(School of Computer and Communications Engineering,Liaoning University of Petroleum and Chemical Technology,Fushun 113001,China)

机构地区：[1]辽宁石油化工大学计算机与通信工程学院,抚顺113001

出　　处：《价值工程》2020年第29期200-201,共2页Value Engineering

摘　　要：本文将机器学习算法应用于网络优化过程,基于网络流量分类提出一种快速准确的路由选路方案,可应用于多种复杂网络环境。基于主成分分析及半监督聚类理论提出了一种基于QoS类别的网络流量分类方案,根据分类结果进行路由选路,选路过程采用Q-Learning算法,通过对Q表的更新进行最佳路径的选择。实验结果表明,该方案具有良好的网络优化效果。In this paper,machine learning algorithm is applied to network optimization process,and a fast and accurate routing scheme based on network traffic classification is proposed,which can be applied to a variety of complex network environments.Based on principal component analysis and semi-supervised clustering theory,a network traffic classification scheme based on QoS categories is proposed.According to the classification results,the routing is selected.Q-learning algorithm is adopted in the routing process,and the optimal path is selected by updating the Q table.The experimental results show that the scheme has a good network optimization effect.

关 键 词：机器学习 Q-LEARNING SDN QoS路由分配 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程]