机构地区: 辽宁师范大学数学系
出 处: 《系统仿真学报》 2004年第3期519-520,524,共3页
摘 要: 支持向量机是一种基于统计学习理论的新颖的机器学习方法,该方法已广泛用于解决分类和回归问题。将支持向量回归算法应用于混沌时间序列预测中,并同BP网络及RBF网络的预测结果进行了比较分析。仿真实验表明,支持向量回归方法具有很好的泛化能力和一定的噪声容忍能力。 Support vector machines (SVM) are a kind of novel machine learning methods based on statistical learning theory, which has been developed to solve classification and regression problems. This paper applies support vector regression (SVR) to chaotic time series prediction, and compares the prediction results with BP network and RBF network. The results of simulation experiments show that SVR has a good generalization ability and capability of tolerating noise.
关 键 词: 支持向量机 支持向量回归 混沌时间序列 核函数
领 域: [自动化与计算机技术] [自动化与计算机技术]