支持向量机回归算法及其应用

支持向量机是建立在统计学习理论基础上的通用学习方法，它可较好地解决以往很多学习方法的小样本、非线性、过学习、高维数、局部极小点等实际问题。笔者利用支持向量回归理论和方法，建立支持向量机的预测模型，并利用winSVM和MATLAB软件进行了实例预测，与二次回归预测值相比较，支持向量机预测模型具有更好的预测精度，且有很强的推广能力。

Support Vector Machine Regression Algorithm and Its Application

Support vector machine(SVM) is a general learning method built on the basis of statistical learning theory,and it can solve a lot of previous practical problems satisfactorily,including small samples,non-linear,over-learning,high-dimension,local minimum point,and so on.In this article,the theory and the method of the support vector regression were used to set up the support vector machine prediction model,and the case study for the purpose of data forecasting was performed using the software called winSVM and MATLAB.Comparing with the second regression forecast,the SVM prediction model has better forecasting accuracy and is worthy of being widely promoted.