摘要
基于统计学习理论的支持向量机(SVM)是一种新型的机器学习方法,描述了SVM在模式识别和回归估计中的基本思想。在大训练样本情况下,用传统的方法求解SVM问题计算复杂,针对该问题探讨了一系列的SVM训练算法,并对其进行了比较。SVM由于其良好的泛化能力和全局最优性能,在模式识别、数据挖掘、非线性系统建模和控制等领域中展现出广泛的应用前景。
Support vector machine based on SLT is a kind of novel machine learning methods. The basic ideas of SVM for pattern recognition and regression are introduced. Under large samples,it is considerable complex to solve SVM questions by traditional methods. A series of training algorithms are discussed and compared. SVM has been applied to many fields such as pattern recognition,data mining, modeling and control of nonlinear system due to good generalization ability and globally optimal performance.
出处
《石油化工自动化》
CAS
2006年第3期34-38,共5页
Automation in Petro-chemical Industry
关键词
支持向量机
模式识别
回归估计
训练算法
support vector machine
pattern recognition
regression
training algorithm
