全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于特征空间变换的纠错输出编码

DOI: 10.13195/j.kzyjc.2014.0843, PP. 1597-1602

Keywords: 纠错输出编码,特征空间,基分类器独立性

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对基于纠错输出编码多类分类中如何保证基分类器差异性的问题,提出一种基于特征空间变换的编码方法.该方法引入特征空间,将编码矩阵扩展成三维矩阵;然后基于二类划分,利用特征变换得到不同的特征子空间,从而训练得到差异性大的基分类器.基于公共数据集的实验结果表明:该方法能够比原始的编码矩阵获得更优的分类性能,同时增加了基分类器的差异性;该方法适用于任何编码矩阵,为大数据的分类提供了新的思路.

References

[1]  Kong E, Dietterich T G. Error correcting output codes corrects bias and variance[C]. Proc of the 21th Int Conf on Machine Learning. California, 1995: 313-321.
[2]  Dietterich T G, Bakiri G Solving. Multi-class learning problems via error-correcting output codes[J]. J of Artificial Intelligence Research, 1995, 34(2):263-286.
[3]  Francesco Ricci, David W Aha. Error correcting output codes for local learners[C]. Proc of the 10th European Conf on Machine Learning. Chemita: Springer, 1998: 280-291.
[4]  Masulli F, Valentini G. Effectiveness of error correcting output coding methods in ensemble and monolithic learning machines[J]. Pattern Analysis and Application, 2003, 65(6): 285-300.
[5]  Matthew Prior, Terry Windeatt. Over-fitting in ensembles of neural network classifiers within ECOC frameworks[J]. Lecture Notes in Computer Science, 2005, 3541(1): 286-295.
[6]  Garcia-Pedrajas N, Ortiz-Boyer D. An empirical study of binary classifier fusion methods for multiclass classification[J]. Information Fusion, 2011, 12(2): 111-130.
[7]  Mohammad Ali Bagheri, Gholam Ali Montazer, Ehsanollah Kabir. A subspace approach to error correcting output codes[J]. Pattern Recogniton Letters, 2013, 34(1): 176-184.
[8]  Mohammad Ali Bagheri, Qigang Gao, Sergio Escalera. A gentic-based subspace analysis method for improving error-correcting output coding[J]. Pattern Recognition, 2013, 46(10): 2830-2839.
[9]  Miguel Angel Bautista, Sergio Escalera, Xavier Baro, et al. On the design of an ecoc-compliant genetic algorithm[J]. Pattern Recogntion, 2014, 47(2): 865-884.
[10]  周进登, 王晓丹. 基于混淆矩阵的自适应纠错输出编码多类分类方法[J]. 系统工程与电子技术, 2012, 34(7): 220-226.
[11]  (Zhou J D,Wang X D. Multiclass classification of adaptive error-correcting output codes based on confusion matrix[J]. Systems Engineering and Electronics, 2012, 34(7): 220-226.)
[12]  Zhou J D, Wang X D, Zhou H J, et al. Decoding design based on conditional probabilities in ternary errorcorrecting output codes[J]. Pattern Recognition, 2012, 45(4): 1342-1351.
[13]  雷蕾, 王晓丹, 罗玺, 等. ECOC 多类分类研究综述[J]. 电子学报, 2014, 42(9): 1794-1800.
[14]  (Lei L, Wang X D, Luo X, et al. An overview of multiclassification based on error-correcting output codes[J]. Acta Electrinica Sinica, 2014, 42(9): 1794-1800.)
[15]  周进登, 王晓丹, 周红建, 等. 基于最小?? 近邻错分率编码确定方法及其在多类分类中的应用[J]. 控制与决策, 2011, 26(9): 1296-1302.
[16]  (Zhou J D,Wang X D, Zhou H J, et al. Designing of output codes based on minimal ?? nearest neighbor classifying error and its application in multi-class classification[J]. Control and Decision, 2011, 26(9): 1296-1302.)
[17]  边肇祺, 张学工. 模式识别[M]. 第2 版. 北京: 清华大学出版社, 1999: 62-65.
[18]  (Bian Z Q, Zhang X G. Pattern recognition[M]. 2nd ed. Beijing: Tsinghua University Press, 1999: 62-65.)

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133