针对支持向量机网络流量预测误差较大的问题，提出一种基于多分类支持向量机的网络流量预测方法。该方法在网络流量数据训练阶段通过数据编码，使多分类支持向量机的输出逼近编码值，在预测阶段通过数据解码，将多分类支持向量机的输出转换为实际的网络流量预测结果，从而有效地降低了预测误差。实验结果显示，该方法的预测结果与实际采集的网络流量数据具有相同的变化趋势；在同等实验条件下，该方法预测结果的均方根误差为0.487，而单一支持向量机方法、BP 神经网络方法预测结果的均方根误差分别为1.0954 和2.3642，表明基于多分类支持向量机的网络流量预测方法具有更高的准确性。
Support vector machines (SVMs) have been used for network traffic prediction, but there often exist large prediction errors. This paper presents a network traffic prediction method based on multi-class support vector machine. Through establishing an encoding method in the training phase, the training output is binary encoded, and is established by one to one correspondence with the training input, training the multi-class support vector machine. By constructing a decoding method in the prediction phase, the prediction output is binary decoded, approaching the real value, realizing network traffic prediction and reducing the prediction errors. In simulation experiments, comparison of the real network traffic data and prediction results shows that they have the same evolution trends. Under the same condition, the accuracy of single SVM algorithm is about two times that of BP algorithm, while the accuracy of the proposed algorithm is about two times that of single SVM algorithm. The experiment results show that the proposed method has higher prediction accuracy in contrast with existing algorithms such as those of single SVM and BP neural network.
[1] 林楠, 李翠霞. SVM在非线性网络流量预测中的应用研究[J]. 计算机真, 2011, 28(5): 259-262. Lin Nan, Li Cuixia. Study on nonlinear network traffic prediction based on support vector machine[J]. Computer Simulation, 2011, 28(5): 259-262.
[2] Chen M. Short-term forecasting model of web traffic based on genetic algorithm and neural network[C]//2011 2nd International Conference on Artificial Intelligence. Las Vegas, Nevada, USA: Management Science and Electronic Commerce, 2011: 623-626.
[3] 白志中. 采用支持向量机的网络流量预测研究[J]. 计算机与信息技术, 2009, 193(10): 45-49. Bai Zhizhong. Research on network traffic prediction using support vector machine[J]. Computer & Information Technology, 2009,193(10): 45-49.
[4] 柴佳林, 简银, 刘兴伟. 一种基于支持向量机的IP网络流量预测方法[J]. 西华大学学报: 自然科学版, 2010, 29(1): 54-57. Chai Jiailin, Jian Yin, Liu Xingwei. Prediction of ip network traffic based on support vector machine[J]. Journal of Xihua University: Natural Science Edition, 2010, 29(1): 54-57.
[5] 张颖璐. 基于遗传算法优化支持向量机的网络流量预测[J]. 计算机科, 2008, 35(5): 177-181. Zhang Yinglu. Internet traffic forecasting based on support vector machine optimized by genetic algorithm[J]. Computer Science, 2008, 35 (5): 177-181.
[6] 闵洁, 李潇. 基于最小二乘支持向量机的网络流量预测[J]. 科技创新导, 2010, 3(1): 206-207. Min Jie, Li Xiao. Internet traffic forecasting based on least squares support vector machine[J]. Science and Technology Innovation Herald, 2010, 3(1): 206-207.
[7] 曾勍炜, 徐知海, 付爱英, 等. 融合蚁群算法与支持向量机的网络流量测[J]. 南昌大学学报, 2011, 35(4): 406-408. Zeng Qingwei, Xu Zhihai, Fu Aiying, et al. Prediction of network traffic by combination of ant colony optimization and support vector machine[J]. Journal of Nanchang University, 2011, 35(4): 406-408.
[8] 苟博, 黄贤武. 支持向量机多类分类方法[J]. 数据采集与处理, 2006, 21 (3): 334-339. Gou Bo, Huang Xianwu. Svm multi-class classification on[J]. Journal of Data Acquisition & Processing, 2006, 21(3): 334-339.
[9] Tan Y, Qin X, Jia Z, et al. An improved method of traffic forecasting based on Tariff-SASVR[C]//2009 Fifth International Conference on Natural Computation. Tianjin, China: IEEE Computer Society, 2009: 463-467.
[10] Li X, Lu J, Ding L, et al. Building cooling load forecasting model based on LS-SVM[C]//2009 Asia-Pacific Conference on Information Processing. Shenzhen, China: Information Processing APCIP, 2009: 55-58.
[11] Wang J, Peng Y, Peng X. Mobile communication traffic forecast based on a new fuzzy model[C]//IEEE International Instrumentation and Measurement Technology Conference I2MTC'09 IEEE. Singapore: Instrumentation and Measurement Technology Conference, 2009: 872-877.
[12] Asif M T, Dauwels J, Goh C Y, et al. Unsupervised learning based performance analysis of n-support vector regression for speed prediction of a large road network[C]//2012 15th International IEEE Conference on Intelligent Transportation Systems. Anchorage, USA: Intelligent Transportation Systems (ITSC), 2012: 983-988.
[13] Wei X. Supporting vector-machine prediction of network traffic[C]//2011 International Conference on Electrical and Control Engineering. Yichang, China: Electrical and Control Engineering (ICECE), 2011: 3203-3206.
[14] Wang R, Kwong S, Chen D. A new method for multi-class support vector machines by training least number of classifiers[C]//Proceedings of the 2011 International Conference on Machine Learning and Cybernetics. Melbourne, Australia: Machine Learning and Cybernetics (ICMLC), 2011: 648-659.
