摘要
BP神经网络和用于回归的支持向量机(SVR)在非线性回归中表现出很好的学习和预测能力。本文对这两种方法的算法思想进行分析比较,并通过仿真实例对它们的回归性能加以比较,理论和实验结果表明SVR方法在稳定性和泛化性上优于BP网络方法。
Error-backpropagation artificial neural network and Support Vector Machine for regression have shown good learning and forecasting performance in nonlinear regression.The paper analyses and compares algorithm theory and the ability of regression through simulation of both methods.The results indicate that SVR-based method is better than BP network based method in stability and generality.
出处
《安庆师范学院学报（自然科学版）》
2011年第2期94-96,共3页
Journal of Anqing Teachers College(Natural Science Edition)
基金
安徽省教育厅自然科学重点项目(KJ2010A232)资助
