基于ELM-LSSVM的网络流量预测

为了对网络流量进行准确预测，针对传统极限学习机的“过拟合”不足，提出一种极限学习机和最小二乘支持向量机相融合的网络流量预测模型（ELM-LSSVM）。该模型通过相空间重构获得网络流量的学习样本，引入最小二乘支持向量机对极限学习进行改进，并对网络流量训练集进行学习，采用仿真实验对模型性能进行测试。结果表明，ELM-LSSVM提高了网络流量的预测精度，实现了网络流量准确预测，并具有较强的实际应用价值。

Network traffic prediction based on Extreme Learning Machine and Least Square Support Vec-tor Machine

In order to improve the prediction accuracy, aiming at the defects of the over fitting in extreme learning machine, this paper proposes a novel network traffic prediction model based on Extreme Learning Machine and Least Square Support Vector Machine（ELM-LSSVM）. The phase space reconstruction is used to build learning samples of network flow and then the training samples are input to ELM and are learnt in which the Least Squares Support Vector Machine are introduced into Extreme Learning Machine. The simulation experiment is carried out to test the performance. The results show that the proposed model has improved the prediction accuracy of network traffic and has strong practical application value.