A Novel Contrast Pattern Selection Method for Class Imbalance Problems
- 909 Downloads
Selecting contrast patterns is an important task for pattern-based classifiers, especially in class imbalance problems. The main reason is that the contrast pattern miners commonly extract several patterns with high support for the majority class and only a few patterns, with low support, for the minority class. This produces a bias of classification results toward the majority class, obtaining a low accuracy for the minority class. 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.
KeywordsSupervised classification Pattern selection Contrast patterns Imbalanced databases
This work was partly supported by National Council of Science and Technology of Mexico under the scholarship grant 370272.
- 1.Alcalá-Fdez, J., Fernández, A., Luengo, J., Derrac, J., García, S.: KEEL data-mining software tool: data set repository, integration of algorithms and experimental analysis framework. J. Multiple-Valued Logic Soft Comput. 17(2–3), 255–287 (2011)Google Scholar
- 4.Coenen, F., Leng, P.: An evaluation of approaches to classification rule selection. In: Fourth IEEE International Conference on Data Mining, pp. 359–362 (2004)Google Scholar
- 7.Dong, G., Bailey, J.: Contrast Data Mining: Concepts, Algorithms, and Applications. Chapman and Hall/CRC, 1st edn. (2012)Google Scholar
- 14.Kundu, G., Islam, M., Munir, S., Bari, M.: ACN: an associative classifier with negative rules. In: Proceedings of the 11th IEEE International Conference on Computational Science and Engineering, pp. 369–375. IEEE Xplore Press (2008)Google Scholar
- 15.Li, W., Han, J., Pei, J.: CMAR: accurate and efficient classification based on multiple class-association rules. In: Proceedings of the International Conference on Data Mining, ICDM 2001, pp. 369–376. IEEE (2001)Google Scholar
- 16.Liu, B., Hsu, W., Ma, Y.: Integrating classification and association rule mining. In: Proceedings of the Fourth International Conference on Knowledge Discovery and Data mining, KDD 1998, pp. 80–86. AAAI (1998)Google Scholar
- 17.Loyola-González, O., Garcia-Borroto, M., Martínez-Trinidad, J.F., Carrasco-Ochoa, J.A.: An empirical comparison among quality measures for pattern based classifiers. Intell. Data Anal. 18, S5–S17 (2014)Google Scholar
- 24.Wang, Y.J., Xin, Q., Coenen, F.: A novel rule weighting approach in classification association rule mining. In: Seventh IEEE International Conference on Data Mining Workshops, pp. 271–276 (2007)Google Scholar
- 26.Yin, X., Han, J.: CPAR: classification based on predictive association rules. In: Proceedings of the Third SIAM International Conference on Data Mining, SDM 2003, pp. 331–335. SIAM (2003)Google Scholar
- 27.Zhang, X., Dong, G.: Overview and Analysis of Contrast Pattern Based Classification. In: Dong, G., Bailey, J. (eds.) Contrast Data Mining: Concepts, Algorithms, and Applications, Chap. 11. Data Mining and Knowledge Discovery Series, pp. 151–170. Chapman and Hall/CRC (2012)Google Scholar