Margin setting algorithm for pattern classification via spheres

Abstract

Margin setting algorithm (MSA) is a new sphere-based classification algorithm. It employs an artificial immune system approach to construct a number of hyperspheres that cover each class of a given set of data. To gain insights into the classification performance of MSA, it is the first work to analyze two important fundamental problems of MSA as a sphere-based classifier. First, single sphere or multiple spheres are needed to achieve good classification performance in MSA? This problem was presented as sphere analysis, which was experimentally carried out on simulation data sets using Monte Carlo method. The results demonstrated that MSA employs a multiple-sphere strategy instead of one-sphere strategy as its decision boundaries. This strategy allows MSA to achieve lower probabilities of classification error rate. Second, how to adapt the location and size of the hypersphere to achieve good classification performance? This problem was presented as adaption analysis, which was experimentally carried out on real-world data sets compared to the support vector machine and the artificial neural network. The results demonstrated that MSA employs an artificial immune system approach to optimize the locations of the hyperspheres and to shrink the radius of the hypersphere in a certain range using margin as an algorithm parameter. Overall, computational results indicate the advantages of MSA in classification performance.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

References

  1. 1.

    Takigawa I, Kudo M, Nakamura A (2009) Convex sets as prototypes for classifying patterns. Eng Appl Artif Intell 22(1):101–108

    Google Scholar 

  2. 2.

    Takigawa I, Kudo M, Nakamura A (2005) The convex subclass method: combinatorial classifier based on a family of convex sets. In: International workshop on machine learning and data mining in pattern recognition. Springer, pp 90–99

  3. 3.

    Cooper PW (1962) The hypersphere in pattern recognition. Inf Control 5(4):324–346

    MATH  Google Scholar 

  4. 4.

    Cooper PW (1966) A note on an adaptive hypersphere decision boundary. IEEE Trans Electron Comput 6:948–949

    MATH  Google Scholar 

  5. 5.

    Batchelor BG (1978) Classification and data analysis in vector spaces. In: Batchelor BG (ed) Pattern recognition. Springer, Berlin, pp 65–116

    Google Scholar 

  6. 6.

    Sui C, Kwok NM, Ren T (2011) A restricted coulomb energy (rce) neural network system for hand image segmentation. In: 2011 Canadian conference on computer and robot vision. IEEE, pp 270–277

  7. 7.

    Scofield CL, Reilly DL, Elbaum C, Cooper LN (1988) Pattern class degeneracy in an unrestricted storage density memory. In: Neural information processing systems, pp 674–682

  8. 8.

    Lee Y (1989) Classifiers: adaptive modules in pattern recognition systems. Massachusetts Institute of Technology

  9. 9.

    Marchand M, Shawe-Taylor J (2002) The set covering machine. J Mach Learn Res 3:723–746

    MathSciNet  MATH  Google Scholar 

  10. 10.

    Caulfield HJ, Karavolos A, Ludman JE (2004) Improving optical Fourier pattern recognition by accommodating the missing information. Inf Sci 162(1):35–52

    MathSciNet  MATH  Google Scholar 

  11. 11.

    Wang J, Neskovic P, Cooper LN (2005) Pattern classification via single spheres. In: International conference on discovery science. Springer, pp 241–252

  12. 12.

    Hao P-Y, Lin Y-H (2007) A new multi-class support vector machine with multi-sphere in the feature space. In: International conference on industrial, engineering and other applications of applied intelligent systems. Springer, pp 756–765

  13. 13.

    Wang J, Neskovic P, Cooper LN (2006) A minimum sphere covering approach to pattern classification. In: 18th international conference on pattern recognition (ICPR’06). IEEE, pp 433–436

  14. 14.

    Wang J, Neskovic P, Cooper LN (2006) Learning class regions by sphere covering. Brown University, Providence, RI, IBNS Technical Report 2:2006

  15. 15.

    Olmez T (1997) Classification of ECG waveforms by using RCE neural network and genetic algorithms. Electron Lett 33(18):1561–1562

    Google Scholar 

  16. 16.

    Younsi R, Bagnall A (2012) An efficient randomised sphere cover classifier. Int J Data Min Model Manag 4(2):156

    Google Scholar 

  17. 17.

    Younsi R, Bagnall A (2016) Ensembles of random sphere cover classifiers. Pattern Recognit 49:213–225

    Google Scholar 

  18. 18.

    Fu J, Caulfield HJ, Wu D, Tadesse W (2010) Hyperspectral image analysis using artificial color. J Appl Remote Sens 4(1):043514

    Google Scholar 

  19. 19.

    Wang Y, Adhami R, Fu J (2015) A new machine learning algorithm for removal of salt and pepper noise. In: Seventh international conference on digital image processing (ICDIP 2015). International Society for Optics and Photonics, p 96311R

  20. 20.

    Wang Y, Adhmai R, Fu J, Al-Ghaib H (2015) A novel supervised learning algorithm for salt-and-pepper noise detection. Int J Mach Learn Cybern 6(4):687–697

    Google Scholar 

  21. 21.

    Wang Y, Fu J, Adhami R, Dihn H (2016) A novel learning-based switching median filter for suppression of impulse noise in highly corrupted colour images. Imaging Sci J 64(1):15–25

    Google Scholar 

  22. 22.

    Wang Y, Amin MM, Fu J, Moussa HB (2017) A novel data analytical approach for false data injection cyber-physical attack mitigation in smart grids. IEEE Access 5:26022–26033

    Google Scholar 

  23. 23.

    Wang Y, Pan WD, Fu J (2015) Analysis of the margin setting algorithm as a margin-based spherical classification method. In: SoutheastCon 2015. IEEE, pp 1–7

  24. 24.

    Wang Y, Fu J, David Pan W (2018) Impact of setting margin on margin setting algorithm and support vector machine. J Imaging Sci Technol 62(3):30501–30511

    Google Scholar 

  25. 25.

    Schapire RE, Freund Y (2013) Boosting: foundations and algorithms. Kybernetes

  26. 26.

    Qi Z, Meng F, Tian Y, Niu L, Shi Y, Zhang P (2017) Adaboost-llp: a boosting method for learning with label proportions. IEEE Trans Neural Netw Learn Syst 29(8):3548–3559

    Google Scholar 

  27. 27.

    Hofmeyr SA, Forrest S (2000) Architecture for an artificial immune system. Evolut Comput 8(4):443–473

    Google Scholar 

  28. 28.

    Hammersley J (2013) Monte carlo methods. Springer, Berlin

    Google Scholar 

  29. 29.

    Rubinstein RY, Kroese DP (2016) Simulation and the Monte Carlo method, vol 10. Wiley, Hoboken

    Google Scholar 

  30. 30.

    Asuncion A, Newman D (2007) UCI machine learning repository

  31. 31.

    Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20(3):273–297

    MATH  Google Scholar 

  32. 32.

    Kubat M (2015) Artificial neural networks. In: An introduction to machine learning. Springer, Cham, pp 91–111

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Yi Wang.

Ethics declarations

Conflict of interest

All authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Additional information

Publisher's Note

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

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wang, Y., Pan, W.D., Fu, J. et al. Margin setting algorithm for pattern classification via spheres. Pattern Anal Applic (2020). https://doi.org/10.1007/s10044-020-00888-3

Download citation

Keywords

  • Sphere-based classification
  • Margin setting
  • Hypersphere
  • Multi-sphere decision boundary