[1]徐远洋,何志芬,刘 彬.多视图偏多标记分类与标记消歧的联合学习[J].南京师大学报(自然科学版),2025,48(02):74.[doi:10.3969/j.issn.1001-4616.2025.02.008]
　Xu Yuanyang,He Zhifen,Liu Bin.Joint Learning of Multi-View Partial Multi-Label Classification and Label Disambiguation[J].Journal of Nanjing Normal University(Natural Science Edition),2025,48(02):74.[doi:10.3969/j.issn.1001-4616.2025.02.008]
点击复制

多视图偏多标记分类与标记消歧的联合学习()

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
48
期数:
2025年02期
页码:
74
栏目:
计算机科学与技术
出版日期:
2025-04-15

文章信息/Info

Title:
Joint Learning of Multi-View Partial Multi-Label Classification and Label Disambiguation
文章编号:
1001-4616(2025)02-0074-09
作者:
徐远洋; 何志芬; 刘 彬
(南昌航空大学数学与信息科学学院,江西 南昌 330063)
Author(s):
Xu Yuanyang; He Zhifen; Liu Bin
(School of Mathematics and Information Science,Nanchang Hangkong University,Nanchang 330063,China)
关键词:
多视图偏多标记学习; 核空间; 标记消歧; 标记相关性
Keywords:
multi-view partial multi-label learning; kernel space; label disambiguation; label correlation
分类号:
TP181
DOI:
10.3969/j.issn.1001-4616.2025.02.008
文献标志码:
A
摘要:
多视图偏多标记学习主要处理同时具有多个视图和多个相关标记但标记信息不完全准确的数据. 现有多视图偏多标记学习方法大多采用两阶段的方式独立进行标记消歧与多标记分类,其分类性能有待提高. 本文提出了一种多视图偏多标记分类与标记消歧联合学习(joint learning of multi-view partial multi-label classification and label disambiguation,JL-MVPML-LD)框架. 首先,对多视图特征进行多核融合并考虑不同视图的重要性; 其次,自动挖掘实例相关性和标记相关性,并利用它们来促进多视图偏多标记分类和标记消歧的联合学习; 最后,采用交替迭代方法进行求解. 在3个数据集上27种情况下的实验结果验证了本文方法的有效性.
Abstract:
Multi-view partial multi-label learning mainly aims to deal with data with multi-view features and multiple related labels,but the label information is not completely accurate. Most of the existing methods use a two-stage approach to perform label disambiguation and multi-label classification independently,but these methods are sub-optimal. In this paper,a new learning framework of Joint Learning of Multi-View Partial Multi-Label classification and Label Disambiguation(JL-MVPML-LD)is proposed. Firstly,the multi-view features are fused by using the multi-kernel method and the importance of each view is considered. Secondly,the instance correlation and label correlation are mined and learned,and then used to help the joint learning of multi-view partial multi-label classifier and label disambiguation. Finally,JL-MVPML_LD can be solved by alternating solution method. The experimental results in 27 cases on three datasets verify the effectiveness of the proposed method.

参考文献/References:

[1]ZHANG L,ZHANG Q,ZHANG L,et al. Ensemble manifold regularized sparse low-rank approximation for multiview feature embedding[J]. Pattern recognition,2015,48(10):3102-3112.
[2]XIA T,TAO D,MEI T,et al. Multiview spectral embedding[J]. IEEE transactions on systems,man,and cybernetics,part B(Cybernetics),2010,40(6):1438-1446.
[3]XU C,TAO D,XU C. A survey on multi-view learning[J/OL]. arXiv Preprint arXiv:1304.5634,2013.
[4]ZHANG Y,WU J,CAI Z,et al. Multi-view multi-label learning with sparse feature selection for image annotation[J]. IEEE transactions on multimedia,2020,22(11):2844-2857.
[5]LIU B,LI W,XIAO Y,et al. Multi-view multi-label learning with high-order label correlation[J]. Information sciences,2023,624:165-184.
[6]LI Y,ZHANG J,WU H,et al. Consistent and specific multi-view multi-label learning with correlation information[J]. Information sciences,2025,687:121395.
[7]ZHAO D,GAO Q,LU Y,et al. Consistency and diversity neural network multi-view multi-label learning[J]. Knowledge-based systems,2021,218:106841.
[8]YANG Y,XI W,ZHOU L,et al. Rebalanced vision-language retrieval considering structure-aware distillation[J]. IEEE transactions on image processing,2024,33:6881-6892.
[9]LYU G,DENG X,WU Y,et al. Beyond shared subspace:A view-specific fusion for multi-view multi-label learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence,2022,36(7):7647-7654.
[10]ZHONG Q,LYU G,YANG Z. Align while fusion:a generalized nonaligned multiview multilabel classification method[J]. IEEE transactions on neural networks and learning systems,2024:1-10.
[11]YANG Y,GUO J,LI G,et al. Alignment efficient image-sentence retrieval considering transferable cross-modal representation learning[J]. Frontiers of computer science,2024,18(1):181335.
[12]LIU C,WEN J,LUO X,et al. Incomplete Multi-View Multi-Label Learning via Label-Guided Masked View and Category-Aware Transformers[C]//The Thirty-Seventh AAAI Conference on Artificial Intelligence(AAAI-23),2023:8816-8824.
[13]HAO P,LIU K,GAO W. Anchor-guided global view reconstruction for multi-view multi-label feature selection[J]. Information sciences,2024,679:121124.
[14]LYU G,YANG Z,DENG X,et al. L-VSM:Label-driven view-specific fusion for multiview multilabel classification[J]. IEEE transactions on neural networks and learning systems,2024.
[15]YANG Y,BAO R,GUO W,et al. Deep visual-linguistic fusion network considering cross-modal inconsistency for rumor detection[J]. Science china information sciences,2023,66(12):222102.
[16]柴变芳,李政,赵晓鹏,等. 基于泛化图卷积神经网络的深度文档聚类模型[J]. 南京师大学报(自然科学版),2024,(1):82-90.
[17]TAN X,ZHAO C,LIU C,et al. A two-stage information extraction network for incomplete multi-view multi-label classification[C]//Proceedings of the AAAI Conference on Artificial Intelligence,2024,38(14):15249-15257.
[18]LIU C,WEN J,LIU Y,et al. Masked two-channel decoupling framework for incomplete multi-view weak multi-label learning[J]. Advances in neural information processing systems,2023,36:32387-32400.
[19]WEN J,LIU C,DENG S,et al. Deep double incomplete multi-view multi-label learning with incomplete labels and missing views[J]. IEEE transactions on neural networks and learning systems,2023,35(8):11396-11408.
[20]LI A,FENG C,XU S,et al. Graph t-SNE multi-view autoencoder for joint clustering and completion of incomplete multi-view data[J]. Knowledge-based systems,2024,284:111324.
[21]WU J H,WU X,CHEN Q G,et al. Feature-induced manifold disambiguation for multi-view partial multi-label learning[C]//Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining,2020:557-565.
[22]孙林,施恩惠,司珊珊,等. 基于AP聚类和互信息的弱标记特征选择方法[J]. 南京师大学报(自然科学版),2022,45(3):108-115.
[23]LIU X,SUN L,FENG S. Incomplete multi-view partial multi-label learning[J]. Applied intelligence,2022,52(3):3289-3302.
[24]WANG Z,XU Y. A two-stage multi-view partial multi-label learning for enhanced disambiguation[J]. Knowledge-based systems,2024,293:111680.
[25]LIU B Q,JIA B B,ZHANG M L. Towards enabling binary decomposition for partial multi-label learning[J]. IEEE transactions on pattern analysis and machine intelligence,2023,45(11):13203-13217.
[26]ZHANG M L,FANG J P. Partial multi-label learning via credible label elicitation[J]. IEEE transactions on pattern analysis and machine intelligence,2020,43(10):3587-3599.
[27]ZOU Y,HU X,LI P,et al. Learning shared and non-redundant label-specific features for partial multi-label classification[J]. Information sciences,2024,656:119917.
[28]CHEN Z S,WU X,CHEN Q G,et al. Multi-view partial multi-label learning with graph-based disambiguation[C]//Proceedings of the AAAI Conference on artificial intelligence,2020,34(4):3553-3560.
[29]XU N,WU Y D,QIAO C,et al. Multi-view partial multi-label learning via graph-fusion-based label enhancement[J]. IEEE transactions on knowledge and data engineering,2022,35(11):11656-11667.
[30]YUAN J,LIU W,GU Z,et al. A unified framework for graph-based multi-view partial multi-label learning[J]. IEEE access,2023,11:49205-49215.
[31]耿新,徐宁. 标记分布学习与标记增强[J]. 中国科学:信息科学,2018(5):5.

备注/Memo

备注/Memo:
收稿日期:2025-01-20.
基金项目:国家自然科学基金项目(62362051)、江西省自然科学基金项目(2023BAB202047).
通讯作者:何志芬,博士,副教授,研究方向:机器学习、计算机视觉. E-mail:zfhe323@nchu.edu.cn
更新日期/Last Update: 2025-04-15