Advertisement

A Data Representation Approach to Support Imbalanced Data Classification Based on TWSVM

  • C. JimenezEmail author
  • A. M. Alvarez
  • A. Orozco
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11401)

Abstract

Imbalance classification requires to represent the input data adequately to avoid biased results towards the class with the greater samples number. Here, we introduce an enhanced version of the famous twin support vector machine (TWSVM) classifier by incorporating an extended dual formulation of its quadratic programming optimization. Besides a centered kernel alignment (CKA)-based representation is used to avoid data overlapping. In particular, our approach, termed enhanced TWSVM (ETWSVM), allows representing the input samples in a high dimensional space (possibly infinite) after reformulation of the TWSVM dual form. Obtained results for binary classification demonstrate that our ETWSVM can reveal relevant data structures diminishing overlapping and biased classification results under imbalance scenarios. Moreover, ETWSVM notably adopts the lowest computational cost for training in comparison to state-of-the-art methods.

Keywords

TWSVM Imabalanced data Kernel enhacenment 

Notes

Acknowledgments

Under grants support by the project: “Desarrollo de un sistema de posorte clínico basado en el procesamiento estócastico para mejorar la resolusión espacial de la resonancia magnética estructural y de difusión con aplicación al procedimiento de la ablación de tumores”, code: 111074455860, funded by COLCIENCIAS. Moreover, C. Jimenez is partially financed by the project E6-18-09: “Clasificador de máquinas de vectores de soporte para problemas desbalanceados con selección automática de parámetros”, funded by Vicerrectoria de Investigación, innovación y extension and by Maestría en Ingeniería Eléctrica, both from Universidad Tecnológica de Pereira.

References

  1. 1.
    Haixiang, G., et al.: Learning from class-imbalanced data: review of methods and applications. Expert Syst. Appl. 73, 220–239 (2017)CrossRefGoogle Scholar
  2. 2.
    Loyola-González, O., et al.: Study of the impact of resampling methods for contrast pattern based classifiers in imbalanced databases. Neurocomputing 175, 935–947 (2016)CrossRefGoogle Scholar
  3. 3.
    Branco, P., Torgo, L., Ribeiro, R.P.: A survey of predictive modeling on imbalanced domains. ACM Comput. Surv. 49(2), 31 (2016)CrossRefGoogle Scholar
  4. 4.
    García, S., et al.: Dynamic ensemble selection for multi-class imbalanced datasets. Inf. Sci. 445, 22–37 (2018)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Xanthopoulos, P., Razzaghi, T.: A weighted support vector machine method for control chart pattern recognition. Comput. Ind. Eng. 70, 134–149 (2014)CrossRefGoogle Scholar
  6. 6.
    Tang, L., Tian, Y., Yang, C.: Nonparallel support vector regression model and its SMO-type solver. Neural Netw. (2018)Google Scholar
  7. 7.
    Qi, Z., Tian, Y., Shi, Y.: Structural twin support vector machine for classification. Knowl. Based Syst. 43, 74–81 (2013)CrossRefGoogle Scholar
  8. 8.
    Liu, L., et al.: Between-class discriminant twin support vector machine for imbalanced data classification. In: CAC 2017, pp. 7117–7122. IEEE (2017)Google Scholar
  9. 9.
    Xu, Y.: Maximum margin of twin spheres support vector machine for imbalanced data classification. IEEE Trans. Cybern. 47(6), 1540–1550 (2017)CrossRefGoogle Scholar
  10. 10.
    Alvarez-Meza, A., Orozco-Gutierrez, A., Castellanos-Dominguez, G.: Kernel-based relevance analysis with enhanced interpretability for detection of brain activity patterns. Front. Neurosci. 11, 550 (2017)CrossRefGoogle Scholar
  11. 11.
    Anandarup, R., et al.: A study on combining dynamic selection and data preprocessing for imbalance learning. Neurocomputing 286, 179–192 (2018)CrossRefGoogle Scholar
  12. 12.
    Piri, S., Delen, D., Liu, T.: A synthetic informative minority over-sampling (SIMO) algorithm leveraging support vector machine to enhance learning from imbalanced datasets. Decis. Support Syst. 106, 15–29 (2018)CrossRefGoogle Scholar
  13. 13.
    Shao, Y.H., et al.: An efficient weighted lagrangian twin support vector machine for imbalanced data classification. Pattern Recogn. 47(9), 3158–3167 (2014)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Automatics Research GroupUniversidad Tecnologica de PereiraPereiraColombia

Personalised recommendations