摘要
本文提出了基于主动学习的分类融合算法,将度量层输出的分类器融合问题看作二级分类器的设计问题,将SVM主动学习引入二级分类器设计。该算法在有效减少标注代价的同时获得了较高的分类性能。实验证明该算法在分类性能和标注代价两方面都优于传统分类器融合方法。
Classifier combination based on active learning is proposed in this thesis, which deals with the design of classifier combination systems as training a combiner at the aggregation level and introduces SVM active learning into the design of this multi-category decision combiner.This algorithm presented greatly reduces the number of labeled data the classifier system needs in order to active satisfactory performance. Experiments on standard database show that our algorithm performs better than current classifier combination rules when considering both labeling cost an classification accuracy.
出处
《微计算机信息》
北大核心
2007年第01Z期302-303,282,共3页
Control & Automation
基金
黑龙江省教育厅科学技术研究(项目编号:10541248)
关键词
主动学习
分类器
融合算法
Active learning,Classifier,Merging arithmetic