基于SVM与BP的分类与回归比较研究

A Comparative Research on the Classification and Regression Based on SVM and BP

支持向量机是一种基于统计学习理论的机器学习方法,由于其出色的学习性能,已经广泛应用于解决分类与回归问题。对比研究支持向量机和BP神经网络在分类与回归上的异同,通过仿真实验分析两者在测试集上分类与回归的泛化能力,研究表明支持向量机的泛化能力要优于BP神经网络。

Support Vector Machine(SVM)is a new and pop machine-learning method based on statistical learning theory,widely used in solving classification and regression problems due to its excellent learning quality. In this paper,the support vector machines and BP neural network methods are under research. Based on the simulation experiment results on the classification and regression testing collection,the present SVM can obtain higher generalization ability when compared to another and the differences between them are also analyzed theoretically.