期刊文献+

Cost-Sensitive学习的一个新课题 被引量:2

A New Issue in Cost-Sensitive Learning
下载PDF
导出
摘要 Cost-Sensitive学习注重分类过程中的各种代价,特别是误分类代价和属性代价。在Cost-Sensitive学习中还有一种代价对分类的总代价有较大影响,这种代价可称为分类延时代价,即延误分类而造成的代价。包含分类延时代价的Cost-Sensitive学习是Cost-Sensitive学习中一个新的课题,这个新课题的目标是使分类过程中的误分类代价、检查代价及分类延时代价之和达到最小。给出包含一种简单的分类延时代价的Cost-Sensitive学习,提出一个既"摊薄"延时代价又减少浪费检查代价的检查策略。 Cost-Sensitive learning considers a variety of costs in classification. Test costs and misclassification costs are two most important costs. There is another important cost, it can be called classification delay costs, has not been studied deeply. Cost-Sensitive Learning involving classification delay costs is a new issue in Cost-Sensitive learning. The goal of the study on this issue is to minimize the sum of mis-classification costs, test costs and classification delay costs in classification. In this paper, a kind of Cost-Sensitive Learning involving classification delay costs is discussed, and a novel test strategy is pro-posed, which share a classification delay cost by multiple tests and reduce waste ot batch test costs simultaneously.
出处 《广西师范大学学报(自然科学版)》 CAS 北大核心 2007年第4期55-58,共4页 Journal of Guangxi Normal University:Natural Science Edition
基金 国家自然科学基金资助项目(60573010)
关键词 Cost—Sensitive学习 分类延时代价 序列检查策略 批检查策略 Cost-Sensitive learning classification delay costs sequential test strategies batch test strategies
  • 相关文献

参考文献8

  • 1QUINLAN J R. C4.5 : Programs for machine learning[M]. San Mateo;California : Morgan Kaufmann, 1993 : 1-80.
  • 2NUNEZ M. The use of background knowledge in decision tree induction[J]. Machine learning, 1991,6(3):231-250.
  • 3TAN Ming. Cost-sensitive learning of classification knowledge and its applications in robotics[J]. Machine Learning, 1993,13(1) :7-33.
  • 4TURNEY P. Types of cost in inductive concept learning[C]//Workshop on Cost-Sensitive Learning at the Seventeenth International Conference on Machine Learning. California :Stanford University,2000:15-21.
  • 5LING C X, YANG Qiang, WANG Jian-ning, et al. Decision trees with minimal costs[C]//Proceedings of the Twenty-First International Conference on Machine Learning. New York:ACM Press,2004:69.
  • 6CHAI Xiao-yong, DENG Lin, YANG Qiang, et al. Test-cost sensitive naive bayesian classirication[C]//Proceedings of the Fourth IEEE International Conference on Data Mining. Brighton:IEEE Computer Society Press, 2004: 51-58.
  • 7谢科,张辉,陈鹏,庞斌.文本分类系统关键技术[J].广西师范大学学报(自然科学版),2007,25(2):123-126. 被引量:7
  • 8张玉芳,陈剑敏,熊忠阳.一种改进的贝叶斯文本分类方法[J].广西师范大学学报(自然科学版),2007,25(2):206-209. 被引量:7

二级参考文献14

共引文献12

同被引文献24

  • 1凌晓峰,SHENG Victor S..代价敏感分类器的比较研究(英文)[J].计算机学报,2007,30(8):1203-1212. 被引量:35
  • 2STEFANO C D,SANSONE C,VENTO M. To reject or not to reject:that is the question-answer in case of neural classifiers[J]. IEEE Transactions on SMC, 2008,30(1): 84-94.
  • 3LANDGREBE T C W,TAX D M J,PACLiK P,et al. The interaction between classification and reject performance for distance-based reject-option classifiers [J]. Pattern Recognition Letters, 2006,27 (8): 908-917.
  • 4ZHENG En-hui,ZOU Chao,SUN Jian,et al. SVM-based credit card fraud detection with reject cost and class-dependent error cost[C]//Proceedings of the PAKDD' 09 Workshop :Data Mining When Classes are Imbalanced and Errors Have Cost. Rangsit Campus :Thammasat Printing House, 2009 :50-58.
  • 5AHA D. UCI machine learning datasets[OL/DB]. (2009-09-09). http ://archive. ics. uei. edu/ml/datasets.html.
  • 6VAPNIK V N. The nature of statistical learning theory[M]. New York:Springer-Verlag,1999:138-167.
  • 7DOMINGOS P. Metacost :a general method for making classifiers cost-sensitive [C]//Proceedings of the 5th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM Press, 1999:155-164.
  • 8SaNCHEZ D,VILA M A ,CERDA L ,et al. Association rules applied to credit card fraud detection[J]. Expert Systems with Applications, 2009,36 (2) : 3630-3640.
  • 9SRIVASTAVA A,KUNDU A,SURAL S,et al. Credit card fraud detection using hidden Markov model[J]. IEEE Transactions on Dependable and Secure Computing, 2008,5 (1) : 37-48.
  • 10ELKAN C. The foundations of cost-sensitive learning[C]//Proeeedings of the 7th International Joint Conference on Artificial Intelligence. California : AAAI Press, 2001: 937-978.

引证文献2

二级引证文献4

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部