《计算机应用研究》|Application Research of Computers

基于粗糙集理论的协同训练算法

Novel co-training algorithm based on rough sets

免费全文下载 (已被下载 次)  
获取PDF全文
作者 盛小春,岳晓冬
机构 1.江苏理工学院 云计算与智能信息处理常州市重点实验室,江苏 常州 213001;2.上海大学 计算机工程与科学学院,上海 200444;3.同济大学 计算机科学与技术系,嵌入式系统与服务计算教育部重点实验室,上海 201804
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2013)12-3546-05
DOI 10.3969/j.issn.1001-3695.2013.12.007
摘要 为了提升风险决策环境下协同训练的效果, 提出了一种基于粗糙子空间的协同决策算法。首先利用粗糙集属性约简的概念, 将部分标记数据属性空间分解为两差异性较大的粗糙子空间; 在各子空间上训练分类器, 并依据各分类器决策风险代价及隶属度将无标记数据划分为可信、噪声和待定样本。综合两分类器的分类结果, 标注少量可信无标记样本后重复协同训练。从理论上分析了算法性能提升的区间界, 并在UCI数据集上进行实验, 验证了模型的有效性及效率。
关键词 协同训练;属性约简;粗糙集;粗糙子空间;决策风险
基金项目 国家自然科学基金资助项目(61103067)
常州市云计算与智能信息处理重点实验室资助项目(CM20123004)
江苏理工学院青年基金资助项目(KYY11093)
本文URL http://www.arocmag.com/article/01-2013-12-007.html
英文标题 Novel co-training algorithm based on rough sets
作者英文名 SHENG Xiao-chun, YUE Xiao-dong
机构英文名 1. Key Laboratory of Cloud Computing & Intelligent Information Processing of Changzhou City, Jiangsu University of Technology, Changzhou Jiangsu 213001, China; 2. School of Computer Engineering & Science, Shanghai University, Shanghai 200444, China; 3. Key Laboratory of Embedded System & Service Computing of Ministry of Education, Dept. of Computer Science & Technology, Tongji University, Shanghai 201804, China
英文摘要 In order to improve the performance of co-training in the context of decision with risk, this paper proposed a rough subspace-based co-training algorithm. Based on the concept of attribute reduction in rough sets, the algorithm first splitted all condition attributes of partially labeled data into two diverse rough subspaces. Then the two classifiers trained from derived rough subspaces and classified the unlabeled data into confident, noise and uncertain samples with consideration of the classification risk and membership of decision class. Finally, it labeled a few of confident samples for the two classifiers to learn from each other in iterative manner. It theoretically analyzed the performance of proposed algorithm, and empirical results on selected UCI data sets also show its effectiveness.
英文关键词 co-training; attribute reduction; rough sets; rough subspace; decision risk
参考文献 查看稿件参考文献
  [1] BLUM A, MITCHELL T. Combining labeled and unlabeled data with co-training[C] //Proc of the 11th Annual Conference on Computational Learning Theory. New York:ACM Press, 1998:92-100.
[2] 刘君, 熊忠阳, 王银辉. 蛋白质二级结构的协同训练预测方法[J] . 计算机应用研究, 2011, 28(5):1688-1691.
[3] NIGAM K, GHANI R. Analyzing the effectiveness and applicability of co-training[C] //Proc of the 9th ACM International Conference on Information and Knowledge Management. New York:ACM Press, 2000:86-93.
[4] FEGER F, KOPRINSKA I. Co-training using RBF nets and different feature splits[C] //Proc of International Joint Conference on Neural Networks. 2006:1878-1885.
[5] 唐焕玲, 林正奎, 鲁明羽, 等. 一种结合独立性模型与差异评估的co-training改进方案[J] . 计算机研究与发展, 2008, 45(11):1874-1881.
[6] SALAHELDIN A, El GAYAR N. New feature splitting criteria for co-training using genetic algorithm optimization[C] //Proc of the 9th International Conference on Multiple Classifier Systems. Berlin:Springer-Verlag, 2010:22-32.
[7] 王娇, 罗四维, 曾宪华. 基于随机子空间的半监督协同训练算法[J] . 电子学报, 2008, 36(12A):60-65.
[8] YASLAN Y, CATALTEPE Z. Co-training with relevant random subspaces[J] . Neurocomputing, 2010, 73(10-12):1652-1661.
[9] 于重重, 商利利, 谭励, 等. 半监督学习在不平衡样本集分类中的应用研究[J] . 计算机应用研究, 2013, 30(4):1085-1089.
[10] ZHOU Zhi-hua, LI Ming. Tri-training:exploiting unlabeled data using three classifiers[J] . IEEE Trans on Knowledge and Data Engineering, 2005, 17(11):1529-1541.
[11] LI Ming, ZHOU Zhi-hua. Improve computer-aided diagnosis with machine learning techniques using undiagnosed samples[J] . IEEE Trans on Systems, Man and Cybernetics, Part A:System and Humans, 2007, 37(6):1088-1098.
[12] 李华雄, 周献中, 李天瑞, 等. 决策粗糙集理论及其研究进展[M] . 北京:科学出版社, 2011.
[13] 范会联, 仲元昌. 结合互信息的多目标属性约简[J] . 计算机应用研究, 2012, 29(2):490-492.
收稿日期
修回日期
页码 3546-3550
中图分类号 TP182
文献标志码 A