摘要
在邻域风险最小化原则(VRM)中运用模糊K近邻分类器,来提出一种新的定义邻域半径的方法,从而得出一种新的VRM算法.实例证明这一新算法对解决稀疏小样本的分类和回归有着较好的应用.
It is a new fuzzy method to apply K adjacent classification method to defining vicinal radius in vicinal risk minimization (VRM). In this paper, a new fuzzy VRM algorithm is proposed, and three numerical experiments are done to demonstrate the proposed approach's effectiveness in resolving sparse small samples' classification and regression.
出处
《昆明理工大学学报(理工版)》
2007年第6期108-112,共5页
Journal of Kunming University of Science and Technology(Natural Science Edition)
基金
昆明理工大学青年基金(项目编号:2006-29)
关键词
支持向量机
模糊K近邻
support vector machine
fuzzy K- nearest neighbor