基于类属特征的多标签流形学习分类方法

doi:10.12066/j.issn.1007-2861.2132

上海大学学报(自然科学版) ›› 2021, Vol. 27 ›› Issue (3): 525-534.doi: 10.12066/j.issn.1007-2861.2132

基于类属特征的多标签流形学习分类方法

亢浏越, 黄睿(), 孙广玲

上海大学通信与信息工程学院, 上海 200444

收稿日期:2018-12-10 出版日期:2021-06-30 发布日期:2021-06-27
通讯作者: 黄睿 E-mail:huangr@shu.edu.cn
作者简介:黄睿(1976—), 女, 副教授, 博士, 研究方向为模式识别与智能信息处理. E-mail: huangr@shu.edu.cn
基金资助:
上海市自然科学基金资助项目(16ZR1411100)

Label-specific feature-based multi-label manifold learning

KANG Liuyue, HUANG Rui(), SUN Guangling

School of Communication and Information Engineering, Shanghai University, Shanghai 200444, China

Received:2018-12-10 Online:2021-06-30 Published:2021-06-27
Contact: HUANG Rui E-mail:huangr@shu.edu.cn

摘要/Abstract

摘要：

多标签流形学习(multi-label manifold learning, ML$^{2}$)基于特征流形构建标签流形, 将标签逻辑值转换为实数值, 能更好地反映标签相关性, 提高分类性能. 但是, ML$^{2}$ 与多数多标签分类方法一样, 是基于数据的全部特征进行标签预测, 没有考虑不同特征对不同类别标签的鉴别能力. 因此, 提出一种基于类属特征的多标签流形学习分类(label specific feature based multi-label manifold learning, LSF-ML$^{2}$)方法. 首先, 利用标签数据优化类属特征重要度矩阵, 确定类属特征子集; 再将子集的特征流形映射到标签空间, 使标签从离散型变为数值型; 最后, 通过多输出回归实现分类. 实验结果表明, 所提方法性能优于多种多标签分类方法.

关键词: 多标签学习, 分类, 类属特征, 流形学习, 标签相关性

Abstract:

Multi-label manifold learning (ML$^2$) constructs label manifolds based on feature manifolds and converts logical into numeric labels. This can better reflect the correlations between labels and improve classification performance. However, similar to most methods, ML$^2$ is based on all features and ignores different discriminabilities when different features are used to classify different labels. Therefore, a method we call label-specific feature-based multi-label manifold learning (LSF-ML$^2$) is proposed. First, the labels are used to optimise the feature importance matrix, which can determine the subset of label-specific features. Then, the feature manifold of the subset is mapped to the label space so that the logical labels can be converted into numeric labels. Finally, a multi-output regression is applied for classification. Experimental results show that the proposed method outperforms several existing multi-label classification methods.

Key words: multi-label learning, classification, label-specific feature, manifold learning, label correlation

中图分类号:

TP39

亢浏越, 黄睿, 孙广玲. 基于类属特征的多标签流形学习分类方法[J]. 上海大学学报(自然科学版), 2021, 27(3): 525-534.

KANG Liuyue, HUANG Rui, SUN Guangling. Label-specific feature-based multi-label manifold learning[J]. Journal of Shanghai University（Natural Science Edition）, 2021, 27(3): 525-534.

图/表 5

参考文献 23

[1]	龚露鸣, 徐美华, 刘冬军, 等. 基于混合高斯和 HOG+SVM 的行人检测模型[J]. 上海大学学报(自然科学版), 2018,24(3):341-351.
	Gong L M, Xu M H, Liu D J, et al. Novel model of pedestrian detection based on Gaussian mixture model and HOG+SVM[J]. Journal of Shanghai University (Natural Science Edition), 2018,24(3):341-351.
[2]	Wang X, Liu H, Yang Z, et al. Research and implementation of a multi-label learning algorithm for Chinese text classification[C]// International Conference on Big Data Computing and Communications. 2017: 68-76.
[3]	Liu J, Chang W C, Wu Y, et al. Deep learning for extreme multi-label text classification[C]// International ACM SIGIR Conference on Research and Development in Information Retrieval. 2017: 115-124.
[4]	Boutell M R, Luo J, Shen X, et al. Learning multi-label scene classification[J]. Pattern Recognition, 2004,37(9):1757-1771. doi: 10.1016/j.patcog.2004.03.009
[5]	Sun F, Tang J, Li H, et al. Multi-label image categorization with sparse factor representation[J]. IEEE Transactions on Image Processing, 2014,23(3):1028-1037. doi: 10.1109/TIP.2014.2298978
[6]	Trohidis K, Tsoumakas G, Kalliris G, et al. Multi-label classification of music by emotion[J]. EURASIP Journal on Audio, Speech, and Music Processing, 2011,2011(1):1-9.
[7]	Wu B, Zhong E, Horner A, et al. Music emotion recognition by multi-label multi-layer multi-instance multi-view learning[C]// ACM International Conference on Multimedia. 2014: 117-126.
[8]	Read J, Pfahringer B, Holmes G, et al. Classifier chains for multi-label classification[J]. Machine Learning, 2011,85(3):333-359. doi: 10.1007/s10994-011-5256-5
[9]	Tsoumakas G, Katakis I, Vlahavas I. Random $k$-labelsets for multilabel classification[J]. IEEE Transactions on Knowledge and Data Engineering, 2011,23(7):1079-1089. doi: 10.1109/TKDE.2010.164
[10]	Zhang M L, Zhou Z H. ML-KNN: a lazy learning approach to multi-label learning[J]. Pattern Recognition, 2007,40(7):2038-2048. doi: 10.1016/j.patcog.2006.12.019
[11]	Elisseeff A, Weston J. A kernel method for multi-labelled classification[C]// Advances in Neural Information Processing Systems. 2002: 681-687.
[12]	Zhang M L, Zhou Z H. Multilabel neural networks with applications to functional genomics and text categorization[J]. IEEE Transactions on Knowledge and Data Engineering, 2006,18(10):1338-1351. doi: 10.1109/TKDE.2006.162
[13]	Hou P, Geng X, Zhang M L. Multi-label manifold learning[C]// Thirtieth AAAI Conference on Artificial Intelligence. 2016: 1680-1686.
[14]	Zhang Q W, Zhong Y, Zhang M L. Feature-induced labeling information enrichment for multi-label learning[C]// AAAI Conference on Artificial Intelligence. 2018: 4446-4453.
[15]	Liu B, Xu Z, Wu S, et al. Manifold regularized matrix completion for multilabel classification[J]. Pattern Recognition Letters, 2016,80:58-63. doi: 10.1016/j.patrec.2016.04.017
[16]	Liu B, Li Y, Xu Z. Manifold regularized matrix completion for multi-label learning with ADMM[J]. Neural Networks, 2018,101:57-67. doi: 10.1016/j.neunet.2018.01.011
[17]	Zhu Y, Kwok J T, Zhou Z H. Multi-label learning with global and local label correlation[J]. IEEE Transactions on Knowledge and Data Engineering, 2018,30(6):1081-1094. doi: 10.1109/TKDE.69
[18]	Cai Z, Zhu W. Multi-label feature selection via feature manifold learning and sparsity regularization[J]. International Journal of Machine Learning and Cybernetics, 2018,9(8):1321-1334. doi: 10.1007/s13042-017-0647-y
[19]	Huang R, Jiang W, Sun G. Manifold-based constraint Laplacian score for multi-label feature selection[J]. Pattern Recognition Letters, 2018,112:346-352. doi: 10.1016/j.patrec.2018.08.021
[20]	Zhang M L. LIFT: multi-label learning with label-specific features[C]// International Joint Conference on Artificial Intelligence. 2011: 1609-1614.
[21]	Huang J, Li G, Huang Q, et al. Learning label specific features for multi-label classification[C]// IEEE International Conference on Data Mining. 2015: 181-190.
[22]	Tsoumakas G, Katakis I, Vlahavas I. Mining multi-label data[M]. Boston: Springer, 2009: 667-685.
[23]	Zhang M L, Zhou Z H. A review on multi-label learning algorithms[J]. IEEE Transactions on Knowledge and Data Engineering, 2014,26(8):1819-1837. doi: 10.1109/TKDE.2013.39

基于类属特征的多标签流形学习分类方法

Label-specific feature-based multi-label manifold learning

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 23

相关文章 12

编辑推荐

Metrics

本文评价

[1]	赵倩, 吴悦, 刘宗田. 基于关键 $n$-grams 和门控循环神经网络的文本分类模型[J]. 上海大学学报(自然科学版), 2021, 27(3): 544-552.
[2]	黄嘉宝, 朱永华, 周霁婷, 高文靖. 基于卷积神经网络的多肉植物细粒度图像分类[J]. 上海大学学报(自然科学版), 2020, 26(2): 283-291.
[3]	鲍深, 张金艺, 姚维强, 梁滨. 基于峰度系数的行人水平行走步态细化分类算法[J]. 上海大学学报(自然科学版), 2018, 24(4): 564-571.
[4]	刘惠彬, 陈强, 吴飞, 赵毅. 基于卷积神经网络的固定群体中目标人物分类[J]. 上海大学学报(自然科学版), 2017, 23(6): 874-.
[5]	吴健忠, 周旭捷, 严小芳, 王斐, 白晓庆, 李永萍, 王宇华, 周吉峙. 上海城市垃圾分类减量绿色账户制度的成效与优化建议[J]. 上海大学学报(自然科学版), 2016, 22(2): 197-202.
[6]	李若涵1, 张金艺1,2,3, 徐德政2, 陈兴秀1, 徐秦乐2. 运动分类步频调节的微机电惯性测量单元室内行人航迹推算[J]. 上海大学学报(自然科学版), 2014, 20(5): 612-623.
[7]	林合同，龚云路，秦殿刚，冯铁男，王翼飞. 基于关联规则挖掘的蛋白质相互作用的预测[J]. 上海大学学报(自然科学版), 2012, 18(3): 265-270.
[8]	刘军伟，余小清，万旺根，张静，杨薇. 基于改进型BP神经网络的音频多分类[J]. 上海大学学报(自然科学版), 2012, 18(2): 127-131.
[9]	孙广玲沈宙彪陈岳军. 基于广义可信度阈值判别和多分辨率方向梯度直方图的字符预分类[J]. 上海大学学报(自然科学版), 2010, 16(6): 614-619.
[10]	赵洁苑，龚云路，王翼飞. 基于MiRfilter系统的毛果杨miRNA预测[J]. 上海大学学报(自然科学版), 2010, 16(4): 397-403.
[11]	万旺根;常辽豫;余小清;崔滨;刘晗. 音频信息检索研究现状与发展趋势[J]. 上海大学学报(自然科学版), 2007, 13(4): 363-370 .
[12]	张金艺;熊艳爽. 基于安全充分捕获技术的多时钟数字系统测试矢量生成[J]. 上海大学学报(自然科学版), 2007, 13(1): 4-9 .