Skip to main content
SpringerLink
Log in

ELM-MC: multi-label classification framework based on extreme learning machine

  • Original Article
  • Published:
International Journal of Machine Learning and Cybernetics Aims and scope Submit manuscript

Abstract

Multi-label classification methods aim to a class of application problems where each individual contains a single instance while associates with a set of labels simultaneously. In this paper, we formulate a novel multi-label classification method based on extreme learning machine framework, named ELM-MC algorithm. The essence of ELM-MC algorithm is to convert the multi-label classification problem into some single-label classifications, and fully considers the relationship among different labels. After the classification of one label, the associations with next label are applied to update the learning parameters in ELM-MC algorithm. In addition, we design a backup pool for the hidden nodes. It can help to select relatively suitable hidden nodes to the corresponding label classification case. In the simulation part, six famous databases are applied to demonstrate the satisfied classification accuracy 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
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

References

  1. Tsoumakas G, Katakis I, Taniar D (2007) Multi-label classification: an overview. Int J Data Warehous Min 3(3):1–13

    Article  Google Scholar 

  2. Luo F, Guo W, Yu Y, Chen G (2017) A multi-label classification algorithm based on kernel extreme learning machine. Neurocomputing 260:313–320

    Article  Google Scholar 

  3. Zhang ML, Zhou ZH (2014) A review on multi-label learning algorithms. IEEE Transa Knowl Data Eng 26(8):1819–1837

    Article  Google Scholar 

  4. Sun L, Ji S, Ye J (2008) Hypergraph spectral learning for multi-label classification. In: Acm Sigkdd international conference on knowledge discovery & data mining, DBLP

  5. Tai F, Lin HT (2012) Multilabel classification with principal label space transformation. Neural Comput 24(9):2508–2542

    Article  MathSciNet  Google Scholar 

  6. Freund Y, Schapire RE (1995) A desicion-theoretic generalization of online learning and an application to boosting. In: European conference on computational learning theory, Springer, pp 23–37

  7. Cao J, Shang L, Wang J et al (2018) A novel distance estimation algorithm for periodic surface vibrations based on frequency band energy percentage feature. Mech Syst Signal Process 113:222–236

    Article  Google Scholar 

  8. Bhatia K, Jain H, Kar P et al (2015) Sparse local embeddings for extreme multi-label classification. Adv Neural Inf Process Syst 730–738

  9. Yu HF, Jain P, Kar P et al (2013) Large-scale multi-label learning with missing labels. In: International conference on machine learning

  10. Deng Y, Dai Q, Liu R et al (2013) Low-rank structure learning via nonconvex heuristic recovery. IEEE Trans Neural Netw Learn Syst 24(3):383–396

    Article  Google Scholar 

  11. Bi W, Kwok J (2013) Efficient multi-label classification with many labels. In: International conference on machine learning, pp 405–413

  12. Zhang ML, Zhou ZH (2006) Multi-label neural networks with applications to functional genomics and text categorization. IEEE Trans Knowl Data Eng 18(10):1338–1351

    Article  Google Scholar 

  13. Nam J, Kim J, Gurevych I et al (2014) Large-scale multi-label text classification—revisiting neural networks. In: Joint European conference on machine learning and knowledge discovery in databases. Springer, Berlin, Heidelberg

  14. Wei Y, Xia W, Lin M et al (2016) HCP: a flexible CNN framework for multi-label image classification. IEEE Trans Pattern Anal Mach Intell 38(9):1901–1907

    Article  Google Scholar 

  15. Wang J, Yang Y, Mao J et al (2016) CNN-RNN: a unified framework for multi-label image classification. In: IEEE conference on computer vision and pattern recognition (CVPR). IEEE

  16. Huang GB, Zhu QY, Siew CK (2006) Extreme learning machine: theory and applications. Neurocomputing 70(1):489–501

    Article  Google Scholar 

  17. Huang GB (2014) An insight into extreme learning machines: random neurons, random features and kernels. Cogn Comput 6(3):376–390

    Article  MathSciNet  Google Scholar 

  18. Liang NY, Huang GB, Saratchandran P, Sundararajan N (2006) A fast and accurate online sequential learning algorithm for feedforward networks. IEEE Trans Neural Netw 17(6):1411–1423

    Article  Google Scholar 

  19. Zhang H, Yin Y, Zhang S (2016) An improved elm algorithm for the measurement of hot metal temperature in blast furnace. Neurocomputing 174:232–237

    Article  Google Scholar 

  20. Sun X, Xu J, Jiang C et al (2016) Extreme learning machine for multi-label classification. Entropy 18(6):225

    Article  MathSciNet  Google Scholar 

  21. Ferng CS, Lin HT (2011) Multi-label classification with error-correcting codes. In: Asian conference on machine learning, pp 281–295

  22. Rong HJ, Ong YS, Tan AH, Zhu Z (2008) A fast pruned-extreme learning machine for classification problem. Neurocomputing 72(1):359–366

    Article  Google Scholar 

  23. Huang G, Huang GB, Song S, You K (2015) Trends in extreme learning machines: a review. Neural Netw 61:32–48

    Article  Google Scholar 

  24. Wold S, Esbensen K, Geladi P (1987) Principal component analysis. Chemom Intell Lab Syst 2(1–3):37–52

    Article  Google Scholar 

  25. Cao J, Wang W, Wang J, Wang R (2016) Excavation equipment recognition based on novel acoustic statistical features. IEEE Trans Cybern 47(12):4392–4404

    Article  Google Scholar 

  26. Izenman AJ (2013) Linear discriminant analysis. In: Modern multivariate statistical techniques. Springer texts in statistics. Springer, New York, NY. https://doi.org/10.1007/978-0-387-78189-1_8

    Chapter  MATH  Google Scholar 

  27. Xanthopoulos P, Pardalos PM, Trafalis TB (2013) Linear discriminant analysis. In: Robust data mining. Springer briefs in optimization. Springer, New York, NY. https://doi.org/10.1007/978-1-4419-9878-1_4

    Chapter  MATH  Google Scholar 

  28. Aha DW, Kibler D, Albert MK (2013) Instance-based learning algorithms. Mach Learn 6(1):37–66

    Google Scholar 

  29. Salzberg SL (1994) Book Review: “C4.5: Programs for Machine Learning” by J. Ross Quinlan. Mach Learn 16(3):235–240

    MathSciNet  Google Scholar 

  30. John GH, Langley P (2013) Estimating continuous distributions in Bayesian classifiers. In: Proceedings of the eleventh conference on uncertainty in artificial intelligence

  31. Zeng ZQ, Yu HB, Xu HR, et. al (2008) Fast training support vector machines using parallel sequential minimal optimization. In: 3rd international conference on intelligent system andknowledge engineering, Xiamen, pp 997–1001. IEEE

  32. Rodriguez-Fdez I, Canosa A, Mucientes M et al (2015) STAC: A web platform for the comparison of algorithms using statistical tests. In: 2015 IEEE international conference on fuzzy systems (FUZZ-IEEE), IEEE

  33. Hodges JL, Lehmann EL (1962) Ranks methods for combination of independent experiments in analysis of variance. Ann Math Stat 33:482–497

    Article  MathSciNet  Google Scholar 

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

    Article  Google Scholar 

  35. Goncalves T, Quaresma P (2003) A preliminary approach to the multilabel classification problem of portuguese juridical documents. In: EPIA 2902, Springer, pp 435–444

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 61806208), Tianjin Education Committee Research Project (2018KJ246) and Fundamental Research Funds for the Central Universities 3122018S008).

Author information

Authors and Affiliations

  1. Institute of Applied Artificial Intelligence of the Guangdong-Hong Kong-Macao Greater Bay Area, Shenzhen Polytechnic, Shenzhen, 518055, China

    Haigang Zhang & Jinfeng Yang

  2. Tianjin Key Laboratory for Advanced Signal Processing, Civil Aviation University of China, Tianjin, 300300, China

    Guimin Jia

  3. Basic Experimental Center, Civil Aviation University of China, Tianjin, 300300, China

    Shaocheng Han

  4. School of Computer Science and Engineering, Central South University, Changsha, 410083, China

    Xinran Zhou

Authors
  1. Haigang Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jinfeng Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Guimin Jia
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Shaocheng Han
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Xinran Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jinfeng Yang.

Additional information

Publisher's Note

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

Appendix

Appendix

Here we present the detailed introduction of the problem transformation methods subject to multi-classification applications. Suppose there are four songs that belong to one or more of four classes: New Age, R&B, Country Music and Hip-Hop. Their detailed attributes are shown in Table 5.

Table 5 Data distribution
Full size table

PT3 transformation method creates new categories based on the label attributes of different samples, such that one sample corresponds to one label as shown in Table 6. PT3 method would increase the number of labels with relatively fewer samples. The detailed applications of PT3 method can be referred in [34]. PT4 transformation method splits the multi-label classification problem into many binary classifications based on the number of labels. As shown in Table 7. PT4 copies the original dataset into 4 same dataset labelled by different attributes. One of the negative aspects of PT4 method is requiring relatively many classifiers, of which the number is equal to the number of labels. PT4 method is applied in [35]. PT5 method decomposes each sample into many individuals and each individual is marked by one label. Table 8 presents the transformation results based on PT5 method. Then a special classifier is applied to the transformed dataset, which would output the probabilities of each sample belonging to each label. A threshold is set in advance to determine the final labels of each sample.

Table 6 PT3
Full size table
Table 7 PT4
Full size table
Table 8 PT5
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, H., Yang, J., Jia, G. et al. ELM-MC: multi-label classification framework based on extreme learning machine. Int. J. Mach. Learn. & Cyber. 11, 2261–2274 (2020). https://doi.org/10.1007/s13042-020-01114-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s13042-020-01114-6

Keywords

Search

Navigation