非均衡数据情形的一种协同正则化多视图半监督学习分类器

doi:10.3969/j.issn.0253-2778.2020.05.007

中国科学技术大学学报 ›› 2020, Vol. 50 ›› Issue (5): 596-604.DOI: 10.3969/j.issn.0253-2778.2020.05.007

非均衡数据情形的一种协同正则化多视图半监督学习分类器

崔文泉，陈伟，程浩洋

中国科学技术大学管理学院统计与金融系，安徽合肥 230026

收稿日期:2019-04-14 修回日期:2019-05-17 接受日期:2019-05-17 出版日期:2020-05-31 发布日期:2019-05-17
通讯作者: 崔文泉
作者简介:崔文泉（通讯作者），男，1964年生，博士/副教授.研究方向：数理统计.E-mail: wqcui@ustc.edu.cn
基金资助:
国家自然科学基金（71873128）资助.

A multi-view based semi-supervised classifier with co-regularization for imbalanced data

CUI Wenquan, CHEN Wei, CHENG Haoyang

Department of Statistics and Finance， School of Management, University of Science and of Technology of China,Hefei 230026, China

Received:2019-04-14 Revised:2019-05-17 Accepted:2019-05-17 Online:2020-05-31 Published:2019-05-17

摘要/Abstract

摘要： 利用多视图学习、流形学习以及协同正则化的多重惩罚处理，对含有大量无标签的类别数据提出一种多视图半监督学习的分类器构造方法.该方法由递归提升的方式对数据进行逐次多视图半监督学习，通过适当的标签化、均衡化处理改进每次集成的学习效率直到稳定.通过最小二乘和多分类SVM研究了新方法的性质，给出泛化误差的一个有意义上界，体现了新方法良好的泛化能力.模拟研究和实证分析显示,在有限样本情形下新方法具有良好的表现.

关键词: 半监督学习, 多视图学习, 协同正则化, 非均衡数据, 集成学习

Abstract: A method of constructing a multi-view semi-supervised learning classifier was presented for manifold learning and multi-puncture processing. The multi-view and semi-supervised learning of the data is achieved through recursive optimization, and appropriate labeling and equalization processing, until the efficiency of learning becomes stable. The properties of this multi-classifier were given, for instance, an upper bound of the generalization error, which showed a good capacity for generalization. Simulation and empirical analysis showed that the new method performs well with small samples.

Key words: semi-supervised learning, multi-view learning, co-regularization, imbalanced data, ensemble learning

中图分类号:

O212.1

崔文泉，陈伟，程浩洋. 非均衡数据情形的一种协同正则化多视图半监督学习分类器[J]. 中国科学技术大学学报, 2020, 50(5): 596-604.

CUI Wenquan, CHEN Wei, CHENG Haoyang. A multi-view based semi-supervised classifier with co-regularization for imbalanced data[J]. Journal of University of Science and Technology of China, 2020, 50(5): 596-604.

参考文献

［1］
ZHU X. Semi-supervised learning literature survey[R]. Madison, WI: Department of Computer Sciences, University of Wisconsin-Madison, 2005.
[2] 周志华.半监督学习中的协同训练风范[C]//机器学习及其应用.北京：清华大学出版社，2007.
[3] BLUM A, MITCHELL T. Combining labeled and unlabeled data with co-training[C]//COLT ’98: Proceedings of the Eleventh Annual Conference on Computational Learning Theory. New York: Association for Computing Machinery, 1998: 92-100.
[4] SUN X Y, GONG D W, ZHANG W. Interactive genetic algorithms with large population and semi-supervised learning[J]. Applied Soft Computing Journal, 2012, 12(9): 3004-3013.
[5] ZHOU D, BOUSQUET O, WESTON J, et al. Learning with local and global consistency[C]// NIPS ’03: Proceedings of the 16th International Conference on Neural Information Processing Systems. Cambridge, MA: MIT Press, 2003: 169-176.
[6] BELKIN M, NIYOGI P, SINDHWANI V, et al. Manifold regularization: A geometric framework for learning from examples[J]. The Journal of Machine Learning Research, 2006, 7: 2399-2434.
[7] OLIVER A, ODENA A, RAFFEL C, et al. Realistic evaluation of deep semi-supervised learning algorithms[C]// NIPS ’18: Proceedings of the 32nd International Conference on Neural Information Processing Systems. Red Hook, NY: Curran Associates Inc, 2018: 3239-3250.
[8] WESTON J, RATLE F, MOBAHI H, et al. Deep learning via semi-supervised embedding[C]// Neural Networks: Tricks of the Trade. Berlin: Springer, 2012: 639-655.
[9] CALMA A, REITMAIER T, SICK B. Semi-supervised active learning for support vector machines: A novel approach that exploits structure information in data[J]. Information Sciences, 2018, 456: 13-33.
[10] DRUGMAN T, PYLKKONEN J, KNESER R. Active and semi-supervised learning in ASR: Benefits on the acoustic and language models[DB/OL]. [2019-04-01]. https://arxiv.org/abs/1903.02852.
[11] MUSLEA I, MINTON S, KNOBLOCK C A. Active + semi-supervised learning = robust multi-view learning[C]// ICML ’02: Proceedings of the Nineteenth International Conference on Machine Learning. San Francisco, CA: Morgan Kaufmann Publishers Inc, 2002: 435-442.
[12] SUN S. A survey of multi-view machine learning[J]. Neural Computing and Applications, 2013, 23(7): 2031-2038.
[13] HOTELLING H. Relations between two sets of variates[C]// Breakthroughs in Statistics. New York: Springer, 1992: 162-190.
[14] SUN S. Multi-view Laplacian support vector machines[C]// Advanced Data Mining and Applications. Berlin: Springer, 2011: 209-222.
[15] MINH H Q, BAZZANI L, MURINO V. A unifying framework in vector-valued reproducing kernel Hilbert spaces for manifold regularization and co-regularized multi-view learning[J]. The Journal of Machine Learning Research, 2016, 17(1): 769-840.
[16] 徐蓉，姜峰，姚鸿勋.流形学习概述[J].智能系统学报,2006,1(1): 44-51.

[17] BROUARD C, D'ALCH-BUC F, SZAFRANSKI M. Semi-supervised penalized output kernel regression for link prediction. ICML ’11: Proceedings of the 28th International Conference on International Conference on Machine Learning. Madison,WI: Omnipress, 2011: 593-600.
[18] MINH H Q, SINDHWANI V. Vector-valued manifold regularization[C]// ICML ’11: Proceedings of the 28th International Conference on Machine Learning. Madison, WI: Omnipress, 2011: 57-64.

[19] MICCHELLI C A, PONTIL M. On learning vector-valued functions[J]. Neural Computation, 2005, 17(1):177-204.
[20] MROUEH Y, POGGIO T, ROSASCO L, et al. Multiclass learning with simplex coding[C]// Advances in Neural Information Processing Systems 25 (NIPS 2012). Lake Tahoe, NEV: Neural Information Processing Systems Foundation, 2012.
[21] VIAENE S, DERRIG R A, DEDENE G. Cost-sensitive learning and decision making for massachusetts pip claim fraud data[J]. International Journal of Intelligent Systems, 2004, 19(12): 1197-1215.
[22] MAJID A, ALI S, IQBAL M, et al. Prediction of human breast and colon cancers from imbalanced data using nearest neighbor and support vector machines[J]. Comput Methods Programs Biomed, 2014, 113(3): 792-808.
[23] WU T T, LANGE K. Multicategory vertex discriminant analysis for high-dimensional data[J]. Annals of Applied Statistics, 2010, 4(4): 1698-1721.
[24] 李晓刚.个人信用风险评估的一种基于XGBoost的集成学习方法[D].合肥：中国科学技术大学，2018.
[25] BARTLETT P L, MENDELSON S. Rademacher and Gaussian complexities: Risk bounds and structural results[J]. The Journal of Machine Learning Research, 2002, 3: 463-482.

（上接第595页）

［22］ POPOV P P, WANG H, POPE S P. Specific volume coupling and convergence properties in hybrid particle/finite volume algorithms for turbulent reactive flows［J］. Journal of Computational Physics, 2015, 294: 110-126.
［23］ BARLOW R S, FRANK J H. Effects of turbulence on species mass fractions in methane/air jet flames［J］ . Symposium (International) on Combustion, 1998, 27: 1087-1095.
［24］ ANKARAN R S, HAWKES E R, CHEN J H, et al. Structure of a spatially developing turbulent lean methane-air Bunsen flame［J］ . Proceedings of the Combustion Institute, 2007, 31: 1291-1298.
［25］ BROWN P N, BYRNE G D, HINDMARSH A C. VODE: a variable coefficient ODE solver［J］ . SIAM Journal on Scientific and Statistical Computing, 1989, 10: 1038-1051.

()
()

[1]	崔文泉，余厚莹，侯晓天. 不均衡数据情形的基于聚焦损失的CGAN的集成分类方法[J]. 中国科学技术大学学报, 2020, 50(7): 968-976.
[2]	刘玮康, 秦晓卫, 卫国. 基于模糊度的半监督自步协同下的微信流业务识别[J]. 中国科学技术大学学报, 2020, 50(1): 29-38.
[3]	何捷舟，刘金平，张五霞，肖文辉，唐朝晖，徐鹏飞. 基于在线自适应极限学习机选择性集成的网络入侵检测[J]. 中国科学技术大学学报, 2019, 49(7): 544-554.
[4]	王子一，商琳. 基于电力负荷曲线的设备识别方法[J]. 中国科学技术大学学报, 2019, 49(2): 100-104.
[5]	崔文泉，余德美，程浩洋. 一种处理非均衡数据的非迭代核逻辑回归方法[J]. 中国科学技术大学学报, 2019, 49(12): 965-973.
[6]	崔文泉，黄禹侨. 高维数据情形下的一种基于随机投影的集成分类方法[J]. 中国科学技术大学学报, 2019, 49(12): 974-984.
[7]	张杰，李琳，朱阁. 基于模拟退火半监督学习的信用预测研究[J]. 中国科学技术大学学报, 2018, 48(6): 447-457.
[8]	朱威，颜仕星，张磊，李国正. 基于真实世界临床数据的失眠病判别分析[J]. 中国科学技术大学学报, 2016, 46(10): 867-873.

非均衡数据情形的一种协同正则化多视图半监督学习分类器

A multi-view based semi-supervised classifier with co-regularization for imbalanced data

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价