基于支持向量机的网络参数模型研究

摘要

使用模糊层次分析法对支持向量机的两个参数进行寻优，并用寻找到的最优参数训练支持向量机，建立网络参数模型。首先使用模糊层次分析法对支持向量机两个参数进行寻优，然后用寻找到的最优参数训练支持向量机，最后建立预测模型，预测网络流量。实验结果表明，该方法不但可以较好地跟踪网络流量变化趋势，使网络流量的预测值与实际值非常接近，而且预测误差变化范围波动小，是一种有效且预测精度高的网络参数模型。%Two parameters of support vector machine (SVM) are optimized with fuzzy analytic hierarchy process,with which SVM is trained to establish the network parameter model and predict the network flow. The experimental results show that the method can track the trend of network traffic and make the predicted value of the network traffic is close to the actual value. The range fluctuation of its prediction error is small,so it is an effective network parameter model with high prediction accuracy.