This page is devoted to show the supplementary materials for the paper entitled "A Novel Contrast Pattern Selection Method for Class Imbalance Problems" published during the 9th Mexican Conference on Pattern Recognition (MCPR2017). In this paper, we introduce a contrast pattern selection method for class imbalance problems. Our proposal selects all the contrast patterns for the minority class and a certain percent of contrast patterns for the majority class. Our experiments performed over several imbalanced databases show that our proposal selects significantly better contrast patterns, obtaining better AUC results, than other approaches reported in the literature.

Results obtained by all the evaluated contrast pattern selection methods, considering all the databases
  • AUC results ->
  • Number of patterns by class ->

Statistical Tests

CD diagrams with a statistical comparison of the classification results by using Friedman's test (as a non-parametric test) and Shaffer static procedure (as a post-hoc procedure) [1]. The post hoc comparisons contain α = 0.05, α = 0.10, and adjusted p-values.

  • Comparing our proposal with different k values. ->
  • Comparing our proposal against other state-of-the-art contrast pattern selection methods. -> ->


[1] J. Derrac, S. García, D. Molina, and F. Herrera, “A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms,” Swarm Evol. Comput., vol. 1, no. 1, pp. 3–18, 2011.