使用支持向量机对非线性可分数据进行分类的基本思想是将样本集映射到一个高维线性空间使其线性可分.本文则基于Jordan曲线定理,提出了一种通用的基于分类超曲面的分类方法,简称HSC分类法,它是通过直接构造分类超曲面,根据样本点关于分类曲面的围绕数的奇偶性进行分类的一种新分类判断算法,与SVM方法相比,不需要考虑使用何种核函数,不需要做升维变换,直接解决非线性分类问题.对数据分类应用的结果说明:HSC可以有效地解决非线性数据的分类问题,并能够提高分类效率和准确度. The basic idea of ​​using SVM to classify non-linear separable data is to map the sample set to a high-dimensional linear space to make it linearly separable.This paper presents a general classification hypersurface based Jordan curve theorem The classification method, referred to as the HSC classification method, is a new classification judgment algorithm that directly classifies a classification hypersurface according to the sample points about the parity of the classification surfaces. Compared with the SVM method, it does not need to consider how to use The kernel function can directly solve the problem of non-linear classification without up-dimensional transformation.The results of data classification show that HSC can effectively solve the classification problem of nonlinear data and improve the classification efficiency and accuracy.