面向类不平衡数据的K近邻偏标记学习算法

doi:10.6040/j.issn.1672-3961.0.2021.318

摘要/Abstract

摘要： 针对于类不平衡的偏标记学习问题,在PL-KNN算法的基础上,提出一种可以较有效处理类不平衡问题的偏标记K近邻学习算法(K-nearest neighbor algorithm for class imbalanced partial label learning,IM-PLKNN),利用Parzen窗估计法在样本的不同类别的近邻上设置不同的权重,使多数类样本权重降低,让属于少数类样本的近邻具有更高的权重,降低将少数类样本误测为多数类样本的概率,提高对少数类样本的识别精度。试验结果表明,IM-PLKNN算法较PL-KNN算法在不同评价指标上均有显著提高,特别是对少数类样本的识别精度有大幅度提高。IM-PLKNN算法可以有效提高类不平衡的偏标记K近邻学习算法对数据集整体的预测性能。

关键词: 偏标记学习, 类不平衡, K近邻分类, Parzen窗估计, 代价敏感策略

中图分类号:

TP391

王丽,于明仟,刘文鹏,周瑜,郑蕊蕊,贺建军. 面向类不平衡数据的K近邻偏标记学习算法[J]. 山东大学学报 (工学版), 2022, 52(3): 18-24.

WANG Li, YU Mingqian, LIU Wenpeng, ZHOU Yu, ZHENG Ruirui, HE Jianjun. K-nearest neighbor based partial label learning algorithm for class imbalanced data[J]. Journal of Shandong University(Engineering Science), 2022, 52(3): 18-24.

参考文献

[1] COUR T, SAPP B, TASKAR B. Learning from partial labels[J]. The Journal of Machine Learning Research, 2011, 12: 1501-1536.
[2] COUR T, SAPP B, JORDAN C, et al. Learning from ambiguously labeled images[C] //Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, Florida, USA:IEEE, 2009: 919-926.
[3] JIN R, GHAHRAMANI Z. Learning with multiple labels[J].Advances in Neural Information Processing Systems, 2002, 3(2):921-928.
[4] COME E, OUKHELLOU L. Learning from partially supervised data using mixture models and belief functions[J]. Pattern Recognition, 2009, 42(3):334-348.
[5] LIU L P, DIETTERICH T G.A Conditional multinomial mixture model for superset label learning[C] //Proceedings of the Advances in Neural Information Processing Systems. Lake Tahoe, Nevada: Curran Associates Inc, 2012:557-565.
[6] LIU L P, DIETTERICH T G. Leainability of the superset label learning problem[C] //Proceedings of the 31st International Conference on Machine Learning. Beijing, China: JMLR, 2014:1629-1637.
[7] HULLERMERIER E, BERINGER J. Learning from ambiguously labeled examples[J]. Intelligent Data Analysis, 2006, 10(5): 419-439.
[8] 周斌斌.基于问题转换的偏标记学习算法研究[D].南京:东南大学自动化系,2017. ZHOU Binbin. Research on partial label learning algorithm[D]. Nanjing: School of Automation, Southeast University, 2017.
[9] ZENG Z, XIAO S, JIA K, et al. Learning by associating ambiguously labeled images[C] //Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.California,USA:IEEE, 2013:708-715.
[10] JIE L, ORABONA F. Learning from candidate labeling sets[C] //Advances in Neural Information Processing Systems 23. Cambridge, USA:MIT Press, 2010:1504-1512.
[11] ZHANG M L, ZHOU B B, LIU X Y. Partial label learning via feature-aware disambiguation[C] //Proceedings of the 22nd ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining. New York, USA: Association for Computing Machinery, 2016:1335-1344.
[12] CHEN Y C, PATEL V M, CHELLAPPA R, et al. Ambiguously labeled learning using dictionaries[J]. IEEE Transactions on Information Forensics and Security, 2014, 9(12): 2076-2088.
[13] GONG C, LIU T, TANG Y, et al. A regularization approach for instance-based superset label learning[J]. IEEE Transactions on Cybernetics, 2017, 48(3): 967-978.
[14] LUO J, ORABONA F. Learning from candidate labeling sets[J]. Advances in Neural Information Processing Systems, 2010, 23(3):295-299.
[15] BEYGELZIMER A, LANGFORD J. The offset tree for learning with partial labels[C] //Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: Association for Computing Machinery, 2009: 129-138.
[16] C(＾overO)ME E, OUKHELLOU L, DENOEUX T, et al. Learning from partially supervised data using mixture models and belief functions[J]. Pattern Recognition, 2009, 42(3): 334-348.
[17] TANG C Z, ZHANG M L. Confidence-rated discriminative partial label learning[C] //Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. San Francisco, California, USA: AAAI Press, 2017: 2611-2617.
[18] ZHOU Y, HE J, GU H. Partial label learning via gaussian processes[J]. Cybernetics, IEEE Transactions on, 2017, 47(12):4443-4450.
[19] YE Zhifei, WEN Yimin, LÜ Baoliang. A survey of imbalanced pattern classification problems[J]. China Association of Artificial Intelligence Transcation Intelligent Systems, 2009, 4(2): 148-156.
[20] LI S, WANG Z, ZHOU G, et al. Semi-supervised learning for imbalanced sentiment classification[J]. Plor, 2011, 4(l):1826-1831.
[21] CHEN E, LIN Y, XIONG H, et al. Exploiting probabilistic topic models to improve text categorization under class imbalance[J].Information Processing & Management, 2011, 47(2): 202-214.
[22] LUSA L, BLAGUS R. The class-imbalance problem for high-dimensional class prediction[C] //Proceedings of 2012 11th International Conference on Machine Learning and Applications. Boca Raton, FL, USA: IEEE, 2012:123-126.
[23] WANG J, ZHANG M L. Towards mitigating the class-imbalance problem for partial label learning[C] //Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. New York, USA: Association for Computing Machinery, 2018: 2427-2436.
[24] 周瑜,顾宏.面向不平衡数据的逻辑回归偏标记学习算法[J].大连理工大学学报,2017,57(2):184-188. ZHOU Yu, GU Hong. Logistic regression biased labeling learning algorithm for imbalanced data[J]. Journal of Dalian University of Technology, 2017, 57(2):184-188.
[25] HASTIE T, TIBSHIRANI R, FRIEDMAN J. The elements of statistical learning[J]. Journal of the Royal Statistical Society, 2004, 167(1):192.
[26] HAN H, WANG W Y, MAO B H. Borderline-smote: a new over-sampling method in imbalanced data sets learning[C] //Proceedings of the 2005 International Conference on Advances in Intelligent Computing: Volume Part I(ICIC'05). Berlin, Germany: Springer-Verlag, 2005: 878-887.
[27] JING P, HEISTERKAMP D R, DAI H K. Adaptive quasiconformal kernel nearest neighbor classification[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2004, 26(5):656-661.
[28] PANIS G, LANITIS A. An overview of research activities in facial age estimation using the FG-NET aging database[C] //Proceedings of Lecture Notes in Computer Science(Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).Zurich, Switzerland: Lecture Notes in Computer Science, 2015, 8926: 737-750.
[29] BRIGGS F, FERN X Z, RAICH R. Rank-loss support instance machines for MIML instance annotation[C] //Proceedings of the ACM SIGKDD International Con-ference on Knowledge Discovery and Data Mining. New York, USA: Association for Computing Machinery, 2012: 534-542.

相关文章 15

[1]	聂秀山,马玉玲,乔慧妍,郭杰,崔超然,于志云,刘兴波,尹义龙. 任务粒度视角下的学生成绩预测研究综述[J]. 山东大学学报 (工学版), 2022, 52(2): 1-14.
[2]	蒋桐雨,陈帆,和红杰. 基于非对称U型金字塔重建的轻量级人脸超分辨率网络[J]. 山东大学学报 (工学版), 2022, 52(1): 1-8, 18.
[3]	胡军,杨冬梅,刘立,钟福金. 融合节点状态信息的跨社交网络用户对齐[J]. 山东大学学报 (工学版), 2021, 51(6): 49-58.
[4]	梁晔,马楠,刘宏哲. 图像依赖的显著图融合方法[J]. 山东大学学报 (工学版), 2021, 51(4): 1-7.
[5]	宗欣露,杜佳圆. 基于多目标驱动人工蜂群算法的疏散仿真模型[J]. 山东大学学报 (工学版), 2021, 51(3): 1-6.
[6]	陶亮,刘宝宁,梁玮. 基于CNN-LSTM 混合模型的心律失常自动检测[J]. 山东大学学报 (工学版), 2021, 51(3): 30-36.
[7]	杨修远,彭韬,杨亮,林鸿飞. 基于知识蒸馏的自适应多领域情感分析[J]. 山东大学学报 (工学版), 2021, 51(3): 15-21.
[8]	傅桂霞,邹国锋,毛帅,潘金凤,尹丽菊. 融合Gabor特征与卷积特征的小样本行人重识别[J]. 山东大学学报 (工学版), 2021, 51(3): 22-29.
[9]	张俊三,程俏俏,万瑶,朱杰,张世栋. MIRGAN: 一种基于GAN的医学影像报告生成模型[J]. 山东大学学报 (工学版), 2021, 51(2): 9-18.
[10]	周风余,顾潘龙,万方,尹磊,贺家凯. 多运动视觉里程计的方法与技术[J]. 山东大学学报 (工学版), 2021, 51(1): 1-10.
[11]	王梅,薛成龙,张强. 基于秩空间差异的多核组合方法[J]. 山东大学学报 (工学版), 2021, 51(1): 108-113.
[12]	谢晓兰,王琦. 一种基于多目标的容器云任务调度算法[J]. 山东大学学报 (工学版), 2020, 50(4): 14-21.
[13]	蔡国永,贺歆灏,储阳阳. 基于空间注意力和卷积神经网络的视觉情感分析[J]. 山东大学学报 (工学版), 2020, 50(4): 8-13.
[14]	成科扬,孙爽,詹永照. 基于背景复杂度自适应距离阈值的修正SuBSENSE算法[J]. 山东大学学报 (工学版), 2020, 50(3): 38-44.
[15]	田枫,李欣,刘芳,李闯,孙小强,杜睿山. 基于多模态子空间学习的语义标签生成方法[J]. 山东大学学报 (工学版), 2020, 50(3): 31-37, 44.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed