生物信息学2017，Vol.15Issue(2)：84-89,6.DOI:10.3969/j.issn.1672-5565.20161019001
高斯混合模型的上采样分析
A new over-sampling algorithm by gaussian mixture model
摘要
Abstract
It's significant to solve the class-imbalance problems which have a serious impact on the performance of standard classifiers in machine learning problems.Over-sampling is a popular method in dealing with classimbalance problems,which attempts to balance the sizes of different classes by generating additional samples for minority class.We propose a new over-sampling algorithm that synthesizes new additional samples for minority classes by the Gaussian mixture model.Comparing with several state-of-art related methods on UCI datasets,the experimental results demonstrate that the proposed over-sampling algorithm can reduce the side effect of the class imbalance and help improve the classification performance.关键词
不平衡学习/支持向量机/高斯混合模型/上采样Key words
Imbalance learning/Support vector machine/Gaussian mixture model/Over-sample分类
信息技术与安全科学引用本文复制引用
沈乐阳,孙廷凯..高斯混合模型的上采样分析[J].生物信息学,2017,15(2):84-89,6.基金项目
国家自然科学基金(61373062,61371040) （61373062,61371040）
