[1]邱成羽,陈 秀,程煜婷,等.使用模糊标签驱动标签松弛的多视角分类算法[J].南京师大学报(自然科学版),2026,49(01):96-107.[doi:10.3969/j.issn.1001-4616.2026.01.010]
　Qiu Chengyu,Chen Xiu,Cheng Yuting,et al.Multi-view Classification Driven by Fuzzy Labels for Label Relaxation[J].Journal of Nanjing Normal University(Natural Science Edition),2026,49(01):96-107.[doi:10.3969/j.issn.1001-4616.2026.01.010]
点击复制

使用模糊标签驱动标签松弛的多视角分类算法()

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

卷:
49
期数:
2026年01期
页码:
96-107
栏目:
计算机科学与技术
出版日期:
2026-02-10

文章信息/Info

Title:
Multi-view Classification Driven by Fuzzy Labels for Label Relaxation
文章编号:
1001-4616(2026)01-0096-12
作者:
邱成羽; 陈 秀; 程煜婷; 谢宇航; 欧哲权; 张远鹏
(南通大学医学信息学系,江苏 南通 226007)
Author(s):
Qiu Chengyu; Chen Xiu; Cheng Yuting; Xie Yuhang; Ou Zhequan; Zhang Yuanpeng
(Department of Medical Informatics,Nantong University,Nantong 226007,China)
关键词:
多视角学习; 模糊聚类; 标签松弛; 机器学习
Keywords:
multi-view learning; fuzzy clustering; label relaxation; machine learning
分类号:
TP391.71
DOI:
10.3969/j.issn.1001-4616.2026.01.010
文献标志码:
A
摘要:
随着数据采集技术的不断发展,多视角数据在医学图像、行为识别与多模态分析等领域得到了广泛应用. 然而,不同视角间的语义差异性与标签获取过程中的主观性,常导致标签噪声和分类鲁棒性下降等问题. 为此,本文提出一种基于模糊标签驱动的标签松弛与一致性监督相结合的多视角分类算法. 该方法通过模糊聚类为每个样本构建软标签表示,以挖掘标签的不确定性与语义模糊性. 随后,在软标签的学习过程中引入视角权重与真实标签的联合约束,引导模型在真实标签与模糊标签之间建立柔性监督机制,实现标签层面的软性过渡. 最终,通过多轮迭代将软标签与视角特征共同优化,学习出具有判别性的特征投影矩阵. 在4个真实世界数据集上得到的结果与其他多视角分类算法以及传统分类方法进行比较,所提出的方法在应对标签噪声与多视角信息融合方面均表现出优越性.
Abstract:
With the continuous advancement of data acquisition technologies,multi-view data has been widely applied in fields such as medical imaging,action recognition,and multimodal analysis. However,semantic discrepancies across different views and subjectivity in the label annotation process often lead to label noise and reduced classification robustness. To address these issues,this paper proposed a multi-view classification algorithm that integrates fuzzy-label-driven label relaxation with consistency regularization. Specifically,fuzzy clustering was employed to construct soft label representations for each sample,aiming to capture the uncertainty and semantic ambiguity in the labels. During the learning of soft labels,the method introduced view-specific weights and joint constraints with ground-truth labels to guide the model in establishing a flexible supervision mechanism between hard and fuzzy labels,thereby achieving a smooth transition at the label level. Finally,through iterative optimization of both soft labels and view features,the model learned a discriminative feature projection matrix. Experimental results on four real-world multi-view datasets demonstrated that the proposed method outperforms conventional multi-view and traditional classification approaches in terms of robustness to label noise and effectiveness in multi-view information integration.

参考文献/References:

[1]Yu Z,Dong Z,Yu C,et al. A review on multi-view learning[J]. Frontiers of Computer Science,2025,19(7):197334.
[2]Xu C,Tao D,Xu C. A Survey on Multi-view Learning[EB/OL]. arXiv,1304.5634,[2013-04-20].
[3]Sun X,Sun S. Adversarial robustness and attacks for multi-view deep models[J]. Engineering Applications of Artificial Intelligence,2021,97:104085.
[4]Shu T,Zhang B,Tang Y Y. Multi-view classification via a fast and effective multi-view nearest-subspace classifier[J]. IEEE Access,2019,7:49669-49679.
[5]Chen Z,Jia Y,Wang S,et al. Image-based methods for automatic identification of elongated and flat aggregate particles[J]. Construction and Building Materials,2023,382:131187.
[6]Yang Y,Li X,Fu J,et al. 3D multi-view squeeze-and-excitation convolutional neural network for lung nodule classification[J]. Medical Physics,2023,50(3):1905-1916.
[7]Liang N,Yang Z,Li L,et al. Label-noise robust classification with multi-view learning[J]. Science China Technological Sciences,2023,66(6):1841-1854.
[8]李超波,李洪均,徐晨. 深度学习在图像识别中的应用[J]. 南通大学学报(自然科学版),2018,17(1):1-9.
[9]Hariri R H,Fredericks E M,Bowers K M. Uncertainty in big data analytics:survey,opportunities,and challenges[J]. Journal of Big Data,2019,6(1):44.
[10]顾苏杭,李晨阳,朱培逸. 人类视觉启发的模糊卷积神经网络[J]. 南通大学学报(自然科学版),DOI:10.12194/j.ntu.20241219002.
[11]Yang Z,Zhang H,Liang N,et al. Semi-supervised multi-view clustering by label relaxation based non-negative matrix factorization[J]. The Visual Computer,2023,39(4):1409-1422.
[12]Fu Z,Duan L,Xie Y,et al. Eeg-based epilepsy identification using multiple feature spaces consistent fusion with label relaxation[J]. Journal of Mechanics in Medicine and Biology,2025,25(02):2540013.
[13]Abdi H,Williams L J,Valentin D. Multiple factor analysis:principal component analysis for multitable and multiblock data sets[J]. WIREs Computational Statistics,2013,5(2):149-179.
[14]Kan M,Shan S,Chen X. Multi-view deep network for cross-view classification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,NV,USA:IEEE,2016:4846-4853.
[15]Tong L,Wu H,Wang M D. Integrating multi-omics data by learning modality invariant representations for improved prediction of overall survival of cancer[J]. Methods,2021,189:74-85.
[16]Witten D M,Tibshirani R,Hastie T. A penalized matrix decomposition,with applications to sparse principal components and canonical correlation analysis[J]. Biostatistics,2009,10(3):515-534.
[17]Chen M,Gao C,Ren Z,et al. Sparse CCA via Precision Adjusted Iterative Thresholding[EB/OL]. arXiv:1311.6186,2013.
[18]Gao C,Ma Z,Zhou H H. Sparse CCA:Adaptive estimation and computational barriers[J]. The Annals of Statistics,2017,45(5):2074-2101.
[19]Zhang Y,Gaynanova I. Joint association and classification analysis of multi-view data[J]. Biometrics,2022,78(4):1614-1625.
[20]Witten D M,Tibshirani R J. Extensions of sparse canonical correlation analysis with applications to genomic data[J]. Statistical Applications in Genetics and Molecular Biology,2009,8(1):28.
[21]Liu X,Zhu X,Li M,et al. Late fusion incomplete multi-view clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2019,41(10):2410-2423.
[22]Liong V E,Nguyen T N T,Widjaja S,et al. AMVNet:assertion-based multi-view fusion network for LiDAR semantic segmentation[Z]. arXiv,2020(2020-12-09).
[23]Wang S,Liu X,Liu L,et al. Late fusion multiple kernel clustering with proxy graph refinement[J]. IEEE Transactions on Neural Networks and Learning Systems,2023,34(8):4359-4370.
[24]Zhang T,Liu X,Gong L,et al. Late fusion multiple kernel clustering with local kernel alignment maximization[J]. IEEE Transactions on Multimedia,2023,25:993-1007.
[25]Zhang Y,Peng S,Xue Z,et al. AMSF:attention-based multi-view slice fusion for early diagnosis of Alzheimer's disease[J]. Peer Journal Computer Science,2023,9:e1706.
[26]Song D,Zhang Z,Li W,et al. Judgment of benign and early malignant colorectal tumors from ultrasound images with deep multi-View fusion[J]. Computer Methods and Programs in Biomedicine,2022,215:106634.
[27]Li Y,Xie X. Two novel deep multi-view support vector machines for multiclass classification[J]. Applied Intelligence,2024,55(3):182.
[28]Zhang N,Ding S,Sun T,et al. Multi-view RBM with posterior consistency and domain adaptation[J]. Information Sciences,2020,516:142-157.
[29]Lecun Y,Bottou L,Bengio Y,et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE,1998,86(11):2278-2324.
[30]Li Y,Nie F,Huang H,et al. Large-scale multi-view spectral clustering via bipartite graph[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Austin,Texas,USA,2015:2750-2756.
[31]Oliva A,Torralba A. Modeling the shape of the scene:a holistic representation of the spatial envelope[J]. International Journal of Computer Vision,2001,42(3):145-175.
[32]Hu X,Liu X,Pedrycz W,et al. Multi-view fuzzy classification with subspace clustering and information granules[J]. IEEE Transactions on Knowledge and Data Engineering,2023,35(11):11642-11655.
[33]You X,Xu J,Yuan W,et al. Multi-view common component discriminant analysis for cross-view classification[J]. Pattern Recognition,2019,92:37-51.
[34]Yang M,Deng C,Nie F. Adaptive-weighting discriminative regression for multi-view classification[J]. Pattern Recognition,2019,88:236-245.
[35]Yu Y,Zhou G,Huang H,et al. A semi-supervised label-driven auto-weighted strategy for multi-view data classification[J]. Knowledge-Based Systems,2022,255:109694.

相似文献/References:

[1]康家银,张文娟.用于图像分割的非局部空间约束的核FCM算法[J].南京师大学报(自然科学版),2019,42(03):122.[doi:10.3969/j.issn.1001-4616.2019.03.016]
　Kang Jiayin,Zhang Wenjuan.Kernelized FCM Algorithm with Non-Local SpatialConstraint for Image Segmentation[J].Journal of Nanjing Normal University(Natural Science Edition),2019,42(01):122.[doi:10.3969/j.issn.1001-4616.2019.03.016]

备注/Memo

备注/Memo:
收稿日期:2025-10-30.
基金项目:国家自然科学基金面上项目(82572382)、江苏高校“青蓝工程”项目、江苏省研究生科研与实践创新计划项目(KYCX24-3561).
通讯作者:张远鹏,博士,教授,研究方向:医学人工智能. E-mail:y.p.zhang@ieee.org
更新日期/Last Update: 2026-02-10