计算机应用研究2017，Vol.34Issue(6)：1659-1662,1667,5.DOI:10.3969/j.issn.1001-3695.2017.06.013
基于流形排序的动态过抽样方法研究
Dynamic over sampling method based on manifold ranking
杨杰明 1乔媛媛 1王林 2曲朝阳 1刘海洋2
作者信息
- 1. 东北电力大学信息工程学院,吉林吉林132012
- 2. 吉林供电公司信息通信分公司,吉林吉林132000
- 折叠
摘要
Abstract
Aiming at the problem that traditional sampling was prone to data redundancy and limited to the processing of static data,this paper proposed a dynamic over sampling method based on manifold ranking.The method used the manifold structure to described data according to the data within the global manifold structure sorting the minority class data.It selected higher sort value to perform a resampling strategy in order to improve the data's equilibrium.The experimental results show that the proposed method can obtain better classification performance than the current similar method,and can effectively improve the recognition performance of the classifier to a small number of classes.关键词
不平衡数据/流形排序/过抽样Key words
unbalanced data/manifold ranking/over sampling分类
信息技术与安全科学引用本文复制引用
杨杰明,乔媛媛,王林,曲朝阳,刘海洋..基于流形排序的动态过抽样方法研究[J].计算机应用研究,2017,34(6):1659-1662,1667,5.
