支持向量机分类回归核函数

基于统计学习理论的支持向量机算法以其优秀的学习性能已广泛用于解决分类与回归问题。分类算法通过求两类样本之间的最大间隔来获得最优分离超平面，其几何意义相当直观，而回归算法的几何意义就不那么直观了。另外，有些适用于分类问题的快速优化算法岁不能用于回归算法中。研究了分类与回归算法之间的关系，为快速分类算法应用于回归模型提供了一定的理论依据。

RESEARCH ON TIlE RELATIONSHIP BETWEEN SUPPORT VECTOR MACHINE CLASSIFICATION AND REGRESSION ALGORITHMS

Support vector machine algorithms based on statistical learning theory have been widely applied to solve classification and regression problems. Classification algorithms get the optimal separating hyperplane by the maximum distance between two samples. The geometric meaning of classification algorithms is very intuitive, but the geometric meaning of regression algorithms is not very intuitive. Additionally, a few fast optimization algorithms suitable for classification algorithms could not be applied to regression models. The relationship between classification and regression algorithms is studied,which can provide definite theory foundation for the application of fast classification algorithms to regression models.