摘要
通过剪枝技术与欠采样技术相结合来选择合适数据,以提高少数类分类精度,研究欠采样技术在不平衡数据集环境下的影响。结果表明,与直接欠采样算法相比,本文算法不仅在accuracy值上有所提高,更重要的是大大改善了g-means值,特别是对非平衡率较大的数据集效果会更好。
This paper proposed pruning and under-sampling combined approaches for selected the representative data as training data to improve the classification accuracy for minority class and investigated the effect of under-sampling methods in the imbalanced class distribution environment. The experimental results show that the accuracy of algorithm of this paper compare with direct undersampling algorithm have increased, the most important is to significantly improve the g-means＇ value. Especially, the effect will be better on the imbalance rate of larger data sets.
出处
《计算机应用研究》
CSCD
北大核心
2012年第3期847-848,共2页
Application Research of Computers
基金
国家自然科学基金资助项目(71071002)
安徽省教育厅自然科学基金资助项目(05010428)
安徽大学人才队伍建设项目
安徽大学学术创新团队项目(KJTD001B)
