一种新的模糊支持向量机算法

A New Fuzzy Support Vectors Machine Algorithm

通过改进的模糊K近邻方法对模糊隶属度进行求解,将求得的隶属度带入模糊支持向量机中。实验表明,采用该方法得出的分类结果与用支持向量机方法和用根据距离求解隶属度的模糊支持向量机方法的结果进行比较,误差最小,而且有效的降低了过学习的问题,证明了该方法的可行性。

In this paper,an improved fuzzy K neighbors method was used to solve the fuzzy membership,which was got into the fuzzy support vector machine。 Experimen show that the classification results obtained by this method,compared with the classification results returned by support vectors machine algorithm and the fuzzy support vectors machine whose membership was solved only based on distance,has the smallest error,and it is effective in reducing the over learning problem,which proved this method is feasible.