|Table of Contents|

Costsensitive boosting algorithms


Research Field:
Publishing date:


Costsensitive boosting algorithms
Li QiujieMao YaobinYe ShuguangWang Zhiquan
School of Automation,NUST,Nanjing 210094,China
boostingcostsensitive boostingcostsensitive learningcostsensitive sampling
In terms of the problem of costsensitive learning,this paper investigates costsensitive extension of boosting.A costsensitive boosting learning framework is proposed based on costsensitive sampling.Through introducing costsensitive sampling in each round of naive boosting,the expectation of costsensitive loss is minimized.Under the above framework,two new costsensitive boosting algorithms are deduced.Meanwhile,issues of the instability existing in early costsensitive boosting algorithms are revealed and explained.Experimental results on UCI(University of California,Irvine)data set and CBCL(Center for Biological & Computational Learning)face data set demonstrate:in terms of the costsensitive classification problem,costsensitive sampling boosting algorithms outperform naive boosting and existing costsensitive boosting algorithms.


[1]闫明松,周志华.代价敏感分类算法的实验比较[J].模式识别与人工智能,2005,18(5):628-635. Yan Mingsong,Zhou Zhihua.An empirical comparative study of costsensitive classification algorithms[J].Pattern Recognition and Artificial Intelligence,2005,18(5):628-635.
[2]Elkan C.The foundations of costsensitive learning[A].Proceedings of the 17th International Joint Conference on Artificial Intelligence:Volume 2[C].San Francisco,CA,USA:Morgan Kaufmann Publishers Inc,2001:973-978.
[3]涂承胜,陆玉昌.Boosting视角[J].计算机科学,2005,32(5):140-143. Tu Chengsheng,Lu Yuchang.The analge of view of boosting[J].Computer Science,2005,32(5):140-143.
[4]Freund Y,Schapire R.A decisiontheoretic generalization of online learning and an application to boosting[J].Journal of Computer and System Sciences,1997,55(1):119-139.
[5]Viola P,Jones M.Rapid object detection using a boosted cascade of simple features[A].Proceedings of IEEE Conference Computer Vision and Pattern Recognition[C].Washington,DC:IEEE Computer Society,2001:511-518.
[6]郭志波,严云洋,杨静宇.基于沃尔什特征和增强型Cascade算法的人脸检测[J].南京理工大学学报,2008,32(1);60-64,72. Guo Zhipo,Yan Yunyang,Yang Jingyu.Fast face detection based on Walsh feature and enhanced cascade algorithm[J].Journal of Nanjing University of Science and Technology,2008,32(1);60-64,72.
[7]Fan W,Stolfo S J,Zhang J,Chan P K.AdaCost:Misclassification costsensitive boosting[A].Proceedings of Sixth International Conference on Machine Learning[C].San Francisco,USA:Morgan Kaufmann Publishers Inc,1999:97-105.
[8]Ting K M.A comparative study of costsensitive boosting algorithm[A].Proceedings of the 17th International Conference on Machine Learning[C].San Francisco,USA:Morgan Kaufmann Publishers Inc,2000:983-990.
[9]Viola P,Jones M.Fast and robust classification using asymmetric AdaBoost and a detector cascade[A].Advances in Neural Information Processing System 14[C].Boston,USA:MIT Press,2001:1311-1318.
[10]Sun Y,Kamel M S,Wong A K C,Wang,Y.Costsensitive boosting for classification of imbalanced data[J].Pattern Recognition.2007,40(12):3358-3378.
[11]MasnadiShirazi H,Vasconcelos N.Costsensitive boosting[J].PAMI,2011,33(2):294-309.
[12]Lozano A C,Abe N.Multiclass costsensitive boosting with pnorm loss functions[A].Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining[C].New York,USA:ACM,2008:506-514.
[13]Friedman J,Hastie T,Tibshirani R.Additive logistic regression:A statistical view of boosting[J].The Annals of Statistics.2000,28(2):337-407.
[14]Mason L,Baxter J,Bartlett P,Frean M.Boosting algorithms as gradient descent[A].Advances in Neural Information Processing Systems 12[C].Boston,USA:MIT Press,2000:512-518.
[15]Friedman J H.Greedy function approximation:A gradient boosting machine[J].The Annals of Statistics,2001,29(5):1189-1232.
[16]Newman D,Hettich S,Blake C,Merz C.UCI repository of machine learning data bases[EB/OL].http://www.ics.uci.edu/~mlearn/MLRepository.html,2007-07-18/2012-02-09.
[17]CBCL FACE DATABASE[EB/OL].http://cbcl.mit.edu/software-datasets/FaceData2.html,2000-02-01/2012-02-09.


Last Update: 2013-02-15