Skip to main content
SpringerLink
Log in

Multi-label classification with weak labels by learning label correlation and label regularization

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

In conventional multi-label learning, each training instance is associated with multiple available labels. Nevertheless, real-world objects usually exhibit more sophisticated properties such as abundant irrelevant features, incomplete labels, noisy labels, as well as class imbalance. Unfortunately, most existing multi-label learning algorithms only discussed one of them and failed to consider the confounding effects of these factors, which will degrade the accuracy of multi-label classification. In this paper, we propose an integrated multi-label learning framework ML-INC that trains the multi-label model while addressing the aforementioned issues simultaneously. Specifically, we first decompose the observed label matrix into an incomplete ground-truth label matrix and a noisy label matrix by employing the low-rank and sparse decomposition scheme. Secondly, a label confidence matrix is learned to supplement the incomplete label matrix by utilizing the high-order label correlation and the label consistency. Additionally, the low-rank structure is adopted to capture the label correlation. Thirdly, a label regularization matrix is introduced to alleviate the effects of class imbalance in the label matrix, and a sparse constraint is imposed on the feature mapping matrix to select relevant discriminative features. Finally, the Alternating Direction Multiplier Method (ADMM) is employed to handle the optimization problem and comprehensive experiments are conducted to certify the effectiveness of the proposed method.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Algorithm 1
Algorithm 2
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14

Similar content being viewed by others

Data Availability

The datasets generated during and/or analysed during the current study are available in the mulan and meka repository, http://mulan.sourceforge.net/datasets and http://meka.sourceforge.net.

Notes

  1. Let \(\mathcal {S}_{i}=\left \{j\lvert y_{ij}=1,j=1,\dots ,q\right \}\) denote the candidate label set of i-th example, which contains the ground-truth labels and noisy labels. The mainly difference between the concept here and [37] is that all ground-truth labels are not required to be in the candidate labels.

  2. http://mulan.sourceforge.net/datasets.html & http://meka.sourceforge.nethttp://meka.sourceforge.net

  3. In MLL, a large-scale dataset usually means the data whose number of instances is more than 5000 [20].

  4. Code: http://palm.seu.edu.cn/zhangml/files/ML-kNN.rar

  5. Code: https://github.com/jiunhwang/Date_and_Code/tree/master/code_and_datahttps://github.com/jiunhwang/Date_and_Code/tree/master/code_and_data

  6. Code: http://www.lamda.nju.edu.cn/files/Glocal.zip

  7. Code: https://github.com/Akbarnejad/ESMCImplementation

  8. Code: https://github.com/John986/Multi-label-Learning-with-Missing-Labelshttps://github.com/John986/Multi-label-Learning-with-Missing-Labels

References

  1. Tsoumakas G, Katakis I, Vlahavas I (2009) Mining multi-label data. In: Data mining and knowledge discovery handbook. Springer, pp 667–685

  2. Zhang ML, Zhou ZH (2013) A review on multi-label learning algorithms. IEEE Trans Knowl Data Eng 26(8):1819–1837

    Article  Google Scholar 

  3. Zhang Y, Wu J, Cai Z, Philip SY (2020) Multi-view multi-label learning with sparse feature selection for image annotation. IEEE Trans Multimed 22(11):2844–2857

    Article  Google Scholar 

  4. Wang R, Ridley R, Qu W, Dai X et al (2021) A novel reasoning mechanism for multi-label text classification. Inf Process Manag 58(2):102441

    Article  Google Scholar 

  5. Cerri R, Barros RC, PLF de Carvalho AC, Jin Y (2016) Reduction strategies for hierarchical multi-label classification in protein function prediction. BMC Bioinf 17(1):1–24

    Article  Google Scholar 

  6. Zhang ML, Wu L (2014) Lift: multi-label learning with label-specific features. IEEE Trans Pattern Anal Mach Intell 37(1):107–120

    Article  Google Scholar 

  7. Zhang ML, Zhou ZH (2007) Ml-knn: a lazy learning approach to multi-label learning. Pattern Recognit 40(7):2038–2048

    Article  MATH  Google Scholar 

  8. Liu XW, Zhu XZ, Li MM, Wang L, Zhu E, Liu TL, Kloft M, Shen DG, Yin JP, Gao W (2019) Multiple kernel k k-means with incomplete kernels. IEEE Trans Pattern Anal Mach Intell 42(5):1191–1204

    Google Scholar 

  9. Huang J, Li GR, Huang QM, Wu XD (2017) Joint feature selection and classification for multilabel learning. IEEE Trans Cybern 48(3):876–889

    Article  Google Scholar 

  10. Huang J, Li GR, Huang QM, Wu XD (2016) Learning label-specific features and class-dependent labels for multi-label classification. IEEE Trans Knowl Data Eng 28(12):3309–3323

    Article  Google Scholar 

  11. Zhu Y, Kwok JT, Zhou ZH (2017) Multi-label learning with global and local label correlation. IEEE Trans Knowl ta Eng 30(6):1081–1094

    Article  Google Scholar 

  12. Teisseyre P (2021) Classifier chains for positive unlabelled multi-label learning. Knowl-Based Syst 213:106709

    Article  Google Scholar 

  13. Huang J, Qin F, Zheng X, Cheng Z, Yuan Z, Zhang W, Huang Q (2019) Improving multi-label classification with missing labels by learning label-specific features. Inform Sci 492:124–146

    Article  MathSciNet  MATH  Google Scholar 

  14. Guan YY, Li WH, Zhang BX, Han B, Ji ML (2021) Multi-label classification by formulating label-specific features from simultaneous instance level and feature level. Appl Intell 51(6):3375–3390

    Article  Google Scholar 

  15. Sun L, Yin TY, Ding WP, Qian YH, Xu JC (2021) Feature selection with missing labels using multilabel fuzzy neighborhood rough sets and maximum relevance minimum redundancy. IEEE Trans Fuzzy Syst 30(5):1197–1211. https://ieeexplore.ieee.org/abstract/document/9333666/

    Article  Google Scholar 

  16. Charte F, Rivera AJ, Del Jesus MJ (2016) Multilabel classification: problem analysis, metrics and techniques. Springer

  17. Boutell MR, Luo J, Shen X, Brown CM (2004) Learning multi-label scene classification. Pattern Recognit 37(9):1757–1771

    Article  Google Scholar 

  18. Elisseeff A, Weston J (2001) A kernel method for multi-labelled classification. In: International conference on neural information processing systems: natural and synthetic, pp 681–687

  19. Sun LJ, Ye P, Lyu GY, Feng SH, Dai GJ, Zhang H (2020) Weakly-supervised multi-label learning with noisy features and incomplete labels. Neurocomputing 413:61–71

    Article  Google Scholar 

  20. Tan AH, Ji XW, Liang JY, Tao YZ, Wu WZ, Pedrycz W (2022) Weak multi-label learning with missing labels via instance granular discrimination. Inf Sci 594:200–216. https://doi.org/10.1016/j.ins.2022.02.011

    Article  Google Scholar 

  21. Zhang J, Li SZ, Jiang M, Tan KC (2020) Learning from weakly labeled data based on manifold regularized sparse model. IEEE Trans Cybern 52(5):3841–3854

    Article  Google Scholar 

  22. Tan AH, Liang JY, Wu WZ, Zhang J (2022) Semi-supervised partial multi-label classification via consistency learning. Pattern Recognit:108839

  23. Sun LJ, Lyu GY, Feng SH, Huang XK (2021) Beyond missing: weakly-supervised multi-label learning with incomplete and noisy labels. Appl Intell 51(3):1552–1564

    Article  Google Scholar 

  24. Sun L, Li MM, Ding WP, Zhang E, Mu XX, Xu JC (2022) Afnfs: adaptive fuzzy neighborhood-based feature selection with adaptive synthetic over-sampling for imbalanced data. Inform Sci 612:724–744. https://doi.org/10.1016/j.ins.2022.08.118

    Article  Google Scholar 

  25. Braytee A, Liu W, Anaissi A, Kennedy PJ (2019) Correlated multi-label classification with incomplete label space and class imbalance. ACM Trans Intell Syst Technol (TIST) 10(5):1–26

    Article  Google Scholar 

  26. Han M, Zhang H (2022) Multiple kernel learning for label relation and class imbalance in multi-label learning. Inform Sci 613:344–356

    Article  Google Scholar 

  27. Bucak SS, Jin R, Jain AK (2011) Multi-label learning with incomplete class assignments. In: CVPR 2011. IEEE, pp 2801–2808

  28. Kong XN, Wu ZM, Li LJ, Zhang RF, Yu PS, Wu H, Fan W (2014) Large-scale multi-label learning with incomplete label assignments. In: Proceedings of the 2014 SIAM international conference on data mining. SIAM, pp 920–928

  29. He Z-F, Yang M, Gao Y, Liu H-D, Yin Y (2019) Joint multi-label classification and label correlations with missing labels and feature selection. Knowl-Based Syst 163:145–158

    Article  Google Scholar 

  30. Wu BY, Jia F, Liu W, Ghanem B, Lyu SW (2018) Multi-label learning with missing labels using mixed dependency graphs. Int J Comput Vis 126(8):875–896

    Article  MathSciNet  MATH  Google Scholar 

  31. Xu M, Jin R, Zhou ZH (2013) Speedup matrix completion with side information: application to multi-label learning. Adv Neural Inf Process Syst 26:2301–2309

    Google Scholar 

  32. Liu B, Li Y, Xu Z (2018) Manifold regularized matrix completion for multi-label learning with admm. Neural Netw 101:57–67

    Article  MATH  Google Scholar 

  33. Tan QY, Yu GX, Domeniconi C, Wang J, Zhang ZL (2018) Multi-view weak-label learning based on matrix completion. In: Proceedings of the 2018 SIAM international conference on data mining, pp 450–458

  34. Wang R, Ye S, Li K, Kwong S (2021) Bayesian network based label correlation analysis for multi-label classifier chain. Inform Sci 554:256–275

    Article  MathSciNet  MATH  Google Scholar 

  35. Vasisht D, Damianou A, Varma M, Kapoor A (2014) Active learning for sparse bayesian multilabel classification. In: Proceedings of the 20th ACM SIGKDD international conference on knowledge discovery and data mining, pp 472–481

  36. Li X, Zhao FP, Guo YH (2015) Conditional restricted boltzmann machines for multi-label learning with incomplete labels. In: Artificial intelligence and statistics. PMLR, pp 635–643

  37. Xie MK, Huang SJ (2018) Partial multi-label learning. In: Proceedings of the AAAI conference on artificial intelligence, pp 6454–6461

  38. Yu GX, Chen X, Domeniconi C, Wang J, Li Z, Zhang ZL, Wu XD (2018) Feature-induced partial multi-label learning. In: 2018 IEEE international conference on data mining (ICDM), pp 1398–1403

  39. Xie MK, Huang SJ (2021) Partial multi-label learning with noisy label identification. IEEE Trans Pattern Anal Mach Intell 44(7):3676–3687. https://doi.org/10.1109/TPAMI.2021.3059290

    Google Scholar 

  40. Li ZW, Lyu GY, Feng SH (2020) Partial multi-label learning via multi-subspace representation. In: IJCAI, pp 2612–2618

  41. Zhang ML, Fang JP (2020) Partial multi-label learning via credible label elicitation. IEEE Trans Pattern Anal Mach Intell 43(10):3587–3599

    Article  Google Scholar 

  42. Liu WW, Wang HB, Shen XB, Tsang I (2021) The emerging trends of multi-label learning. IEEE Trans Pattern Anal Mach Intell. https://doi.org/10.1109/TPAMI.2021.3119334

  43. Weng W, Lin YJ, Wu SX, Li YW, Kang Y (2018) Multi-label learning based on label-specific features and local pairwise label correlation. Neurocomputing 273:385–394

    Article  Google Scholar 

  44. Dembczynski K, Jachnik A, Kotlowski W, Waegeman W, Hüllermeier E (2013) Optimizing the f-measure in multi-label classification: plug-in rule approach versus structured loss minimization. In: International conference on machine learning. PMLR, pp 1130–1138

  45. Guo HX, Li YJ, Shang J, Gu MY, Huang YY, Gong B (2017) Learning from class-imbalanced data: review of methods and applications. Expert Syst Appl 73:220–239

    Article  Google Scholar 

  46. Chen K, Lu BL, Kwok JT (2006) Efficient classification of multi-label and imbalanced data using min-max modular classifiers. In: The 2006 IEEE international joint conference on neural network proceedings. IEEE, pp 1770–1775

  47. Wu BY, Lyu SW, Ghanem B (2016) Constrained submodular minimization for missing labels and class imbalance in multi-label learning. In: Proceedings of the AAAI conference on artificial intelligence, vol 30

  48. Zhang ML, Li YK, Yang H, Liu XY (2020) Towards class-imbalance aware multi-label learning. IEEE Trans Cybern. https://doi.org/10.1109/TCYB.2020.3027509

  49. Xu LL, Wang Z, Shen ZF, Wang YB, Chen EH (2014) Learning low-rank label correlations for multi-label classification with missing labels. In: 2014 IEEE international conference on data mining. IEEE, pp 1067–1072

  50. Wang J, Jebara T, Chang S-F (2008) Graph transduction via alternating minimization. In: Proceedings of the 25th international conference on machine learning, pp 1144–1151

  51. Boyd S, Parikh N, Chu E, Peleato B, Eckstein J (2010) Distributed optimization and statistical learning via the alternating direction method of multipliers. Found Trends Mach Learn 3(1):1–122

    Article  MATH  Google Scholar 

  52. Lin ZC, Liu RS, Su ZX (2011) Linearized alternating direction method with adaptive penalty for low-rank representation. Adv Neural Inf Process Syst 24:612–620

    Google Scholar 

  53. Beck A, Teboulle M (2009) A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM J Imag Sci 2(1):183–202

    Article  MathSciNet  MATH  Google Scholar 

  54. Wang L, Hu JF, Chen CZ (2014) On accelerated singular value thresholding algorithm for matrix completion. Appl Math 5(21):3445

    Article  Google Scholar 

  55. Charte F, Rivera AJ, del Jesus MJ, Herrera F (2015) Addressing imbalance in multilabel classification: measures and random resampling algorithms. Neurocomputing 163:3–16

    Article  Google Scholar 

  56. Hosseini Akbarnejad A, Soleymani Baghshah M (2019) An efficient large-scale semi-supervised multi-label classifier capable of handling missing labels. IEEE Trans Knowl Data Eng 31(2):229–242

    Article  Google Scholar 

  57. Ma ZC, Chen SC (2021) Expand globally, shrink locally: discriminant multi-label learning with missing labels. Pattern Recognit 111:107675

    Article  Google Scholar 

  58. Demšar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7:1–30

    MathSciNet  MATH  Google Scholar 

Download references

Acknowledgements

This work was supported by National Natural Science Foundation of China under Grants 62076221 and 61976194.

Author information

Authors and Affiliations

  1. School of Information Engineering, Zhejiang Ocean University, Zhoushan, Zhejiang, 316022, People’s Republic of China

    Xiaowan Ji, Anhui Tan, Wei-Zhi Wu & Shenming Gu

  2. School of Computer and Information Technology, Shanxi University, Taiyuan, Shanxi, 030006, People’s Republic of China

    Anhui Tan

  3. Key Laboratory of Oceanographic Big Data Mining and Application of Zhejiang Province, Zhoushan, Zhejiang, 316022, People’s Republic of China

    Xiaowan Ji, Anhui Tan, Wei-Zhi Wu & Shenming Gu

Authors
  1. Xiaowan Ji
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Anhui Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Wei-Zhi Wu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Shenming Gu
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Xiaowan Ji: Methodology, Original draft preparation, Investigation, Software, Validation, Review and editing.

Anhui Tan: Conceptualization, Software, Supervision, Review and editing.

Wei-Zhi Wu: Resources, Review and editing.

Shenming Gu: Conceptualization, Supervision, Formal analysis.

Corresponding author

Correspondence to Anhui Tan.

Ethics declarations

Competing of Interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ji, X., Tan, A., Wu, WZ. et al. Multi-label classification with weak labels by learning label correlation and label regularization. Appl Intell 53, 20110–20133 (2023). https://doi.org/10.1007/s10489-023-04562-z

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-023-04562-z

Keywords

Search

Navigation