Skip to main content
SpringerLink
Log in

A relative labeling importance estimation algorithm based on global-local label correlations for multi-label learning

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

In multi-label learning, considering the relative importance between labels can yield better performance than considering the equal importance. To explore relative labeling importance, many existing algorithms introduce the global label correlations. However, the global correlations can only reflect the semantic relation between labels, while ignoring the label correlation differences between different instances. In practical applications, labels with high semantic relevance may not be highly relevant in all instances. In this paper, we consider both global and local label correlations to estimate relative labeling importance. Firstly, we calculate a global label correlation matrix in the whole label space. Secondly, each instance subset is assigned a local label correlation matrix, which is learned from the cosine similarity of labels within the cluster. Based on the assumption that label correlations can be transferred from the original categorical space to the numerical label space, we add global and local label correlation regularization terms. Finally, we integrate the importance estimating and the model training into a unified framework, and propose an alternative optimization algorithm to solve it. To validate the effectiveness of the proposed algorithm, we conduct experiments on thirteen multi-label datasets. Experimental results show that the proposed algorithm outperforms existing multi-label learning algorithms.

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
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Notes

  1. http://www.uco.es/kdis/mllresources/

References

  1. Liu S M, Chen J (2015) A multi-label classification based approach for sentiment classification. Expert Syst Appl 42(3):1083–1093

    Article  Google Scholar 

  2. Deng J, Ren F (2020) Multi-label emotion detection via emotion-specified feature extraction and emotion correlation learning. IEEE Trans Affect Comput PP(99):1–1

    Article  Google Scholar 

  3. Castillo J R, Flores M J (2021) Web-based music genre classification for timeline song visualization and analysis. IEEE Access 9:18801–18816

    Article  Google Scholar 

  4. Wang T, Liu L, Liu N, Zhang H, Zhang L, Feng S (2020) A multi-label text classification method via dynamic semantic representation model and deep neural network. Appl Intell 50(8):2339–2351

    Article  Google Scholar 

  5. de Almeida A M G, Cerri R, Paraiso E C, Mantovani R G, Junior S B (2018) Applying multi-label techniques in emotion identification of short texts. Neurocomputing 320:35–46

    Article  Google Scholar 

  6. Wu B, Lyu S, Hu B, Ji Q (2015) Multi-label learning with missing labels for image annotation and facial action unit recognition. Pattern Recognit 48(7):2279–2289

    Article  Google Scholar 

  7. Ma Y, Xie Q, Liu Y, Xiong S (2020) A weighted knn-based automatic image annotation method. Neural Comput Appl 32(11):6559–6570

    Article  Google Scholar 

  8. Zhang M, Zhang Q, Fang J, Li Y, Geng X (2021) Leveraging implicit relative labeling-importance information for effective multi-label learning. IEEE Trans Knowl Data Eng 33 (5):2057–2070. https://doi.org/10.1109/TKDE.2019.2951561

    Article  Google Scholar 

  9. Hou P, Geng X, Zhang M (2016) Multi-label manifold learning. In: Proceedings of the Thirtieth AAAI conference on artificial intelligence, pp 1680–1686

  10. Zhang Q, Zhong Y, Zhang M (2018) Feature-induced labeling information enrichment for multi-label learning. In: Proceedings of the Thirty-Second AAAI conference on artificial intelligence, pp 4446–4453

  11. He S, Feng L, Li L (2018) Estimating latent relative labeling importances for multi-label learning. In: Proceedings of the IEEE international conference on data mining, pp 1013–1018

  12. Luaces O, D⋅⋅̱aez J, Barranquero J, Coz JJ, Bahamonde A (2012) Binary relevance efficacy for multilabel classification. Progress in Artificial Intelligence 1(4):303–313

  13. Fürnkranz J, Húllermeier E, Mencía EL, Brinker K (2008) Multilabel classification via calibrated label ranking. Mach Learn 73(2):133–153

    Article  MATH  Google Scholar 

  14. Ji Z, Cui B, Li H, Jiang Y, Xiang T, Hospedales T M, Fu Y (2020) Deep ranking for image zero-shot multi-label classification. IEEE Trans Image Process 29:6549–6560

    Article  MathSciNet  MATH  Google Scholar 

  15. Liu H, Chen G, Li P, Zhao P, Wu X (2021) Multi-label text classification via joint learning from label embedding and label correlation. Neurocomputing 460:385–398. https://doi.org/10.1016/j.neucom.2021.07.031

    Article  Google Scholar 

  16. Hua Y, Mou L, Zhu XX (2019) Label relation inference for multi-label aerial image classification. In: Proceedings of the 2019 IEEE international geoscience and remote sensing symposium, pp 5244–5247

  17. Weng W, Lin Y, Wu S, Li Y, Kang Y (2018) Multi-label learning based on label-specific features and local pairwise label correlation. Neurocomputing 273:385–394. https://doi.org/10.1016/j.neucom.2017.07.044

    Article  Google Scholar 

  18. Wang R, Ye S, Li K, Kwong S (2021) Bayesian network based label correlation analysis for multi-label classifier chain. Inf Sci 554:256–275. https://doi.org/10.1016/j.ins.2020.12.010

    Article  MathSciNet  MATH  Google Scholar 

  19. Nazmi S, Yan X, Homaifar A, Doucette E A (2020) Evolving multi-label classification rules by exploiting high-order label correlations. Neurocomputing 417:176–186

    Article  Google Scholar 

  20. Huang SJ, Zhou ZH (2012) Multi-label learning by exploiting label correlations locally. In: Proceedings of the Twenty-Sixth AAAI conference on artificial intelligence, pp 949–955

  21. Huang Z (1997) A fast clustering algorithm to cluster very large categorical data sets in data mining. In: Workshop on research issues on data mining and knowledge discovery, DMKD 1997 in cooperation with ACM SIGMOD’97, Tucson, Arizona, USA, May 11, 1997

  22. Huang Z (1998) Extensions to the k-means algorithm for clustering large data sets with categorical values. Data Min Knowl Disc 2(3):283–304

    Article  Google Scholar 

  23. Bandyopadhyay A, Deb K, Das A, Bag R (2020) Impulse noise removal by k-means clustering identified fuzzy filter: a new approach. Turk J Electr Eng Comput Sci 28

  24. Schubiger M, Banjac G, Lygeros J (2020) GPU Acceleration of ADMM for large-scale quadratic programming. J Parallel Distributed Comput 144:55–67. https://doi.org/10.1016/j.jpdc.2020.05.021

  25. Zhang M, Zhou Z (2014) A review on multi-label learning algorithms. IEEE Trans Knowl Data Eng 26(8):1819–1837

    Article  Google Scholar 

  26. Demiar J, Schuurmans D (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7(1):1–30

    MathSciNet  Google Scholar 

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China (under grants 61976128, 61773247).

Author information

Authors and Affiliations

  1. Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education, School of Computer and Information Technology, Shanxi University, Taiyuan, China

    Yilu Liu & Fuyuan Cao

Authors
  1. Yilu Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Fuyuan Cao
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Fuyuan Cao.

Additional information

Publisher’s note

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

Appendices

Appendix A:: comparison results of label correlations

figure d

Appendix B:: experimental results of convergence

figure e

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, Y., Cao, F. A relative labeling importance estimation algorithm based on global-local label correlations for multi-label learning. Appl Intell 53, 4940–4958 (2023). https://doi.org/10.1007/s10489-022-03700-3

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-022-03700-3

Keywords

Search

Navigation