Soft Computing

, Volume 14, Issue 7, pp 713–728 | Cite as

Multi-objective genetic fuzzy classifiers for imbalanced and cost-sensitive datasets

  • Pietro Ducange
  • Beatrice Lazzerini
  • Francesco Marcelloni
Original Paper

Abstract

We exploit an evolutionary three-objective optimization algorithm to produce a Pareto front approximation composed of fuzzy rule-based classifiers (FRBCs) with different trade-offs between accuracy (expressed in terms of sensitivity and specificity) and complexity (computed as sum of the conditions in the antecedents of the classifier rules). Then, we use the ROC convex hull method to select the potentially optimal classifiers in the projection of the Pareto front approximation onto the ROC plane. Our method was tested on 13 highly imbalanced datasets and compared with 2 two-objective evolutionary approaches and one heuristic approach to FRBC generation, and with three well-known classifiers. We show by the Wilcoxon signed-rank test that our three-objective optimization approach outperforms all the other techniques, except for one classifier, in terms of the area under the ROC convex hull, an accuracy measure used to globally compare different classification approaches. Further, all the FRBCs in the ROC convex hull are characterized by a low value of complexity. Finally, we discuss how, the misclassification costs and the class distributions are fixed, we can select the most suitable classifier for the specific application. We show that the FRBC selected from the convex hull produced by our three-objective optimization approach achieves the lowest classification cost among the techniques used as comparison in two specific medical applications.

Keywords

Genetic fuzzy rule-based classifiers Multi-objective evolutionary algorithms Imbalanced datasets ROC curves Convex hull method 

References

  1. Alcalá R, Gacto MJ, Herrera F, Alcalá-Fdez J (2007) A multi-objective genetic algorithm for tuning and rule selection to obtain accurate and compact linguistic fuzzy rule-based systems. Int J Uncertain Fuzziness Knowl Based Syst 15(5):521–537. doi:10.1142/S0218488507004856 CrossRefGoogle Scholar
  2. Alcalá-Fdez J, Sánchez L, García S, del Jesus MJ, Ventura S, Garrell JM, Otero J, Romero C, Bacardit J, Rivas VM, Fernández JC, Herrera F (2009) KEEL: a software tool to assess evolutionary algorithms to data mining problems. Soft Comput 13(3):307–318. doi:10.1007/s00500-008-0323-y CrossRefGoogle Scholar
  3. Anastasio M, Kupinski M, Nishikawa R (1998) Optimization and FROC analysis of rule-based detection schemes using a multiobjective approach. IEEE Trans Med Imaging 17(10):1089–1093. doi:10.1109/42.746726 CrossRefGoogle Scholar
  4. Antonelli M, Frosini G, Lazzerini B, Marcelloni F (2006) A CAD system for lung nodule detection based on an anatomical model and a fuzzy neural network. In: Proceedings of NAFIPS, Montreal, Canada, 3–6 June, pp 448–453Google Scholar
  5. Asuncion A, Newman D (2007) UCI machine learning repository. University of California, Irvine, School of Information and Computer Sciences. http://www.ics.uci.edu/~mlearn/MLRepository.html
  6. Awai K, Murao K, Ozawa A, Komi M, Hayakawa H, Hori S, Nishimura Y (2004) Pulmonary nodules at chest CT: effect of computer-aided diagnosis on radiologists’ detection performance. Radiology 230(2):347–352. doi:10.1148/radiol.2302030049 CrossRefGoogle Scholar
  7. Batista G, Prati R, Monard M (2004) A study of the behaviour of several methods for balancing machine learning. SIGKDD Explor 6(1):20–29. doi:10.1145/1007730.1007735 CrossRefGoogle Scholar
  8. Casillas J, Cordon O, Del Jesus MJ, Herrera F (2001) Genetic feature selection in a fuzzy rule-based classification system learning process for high-dimensional problems. Inf Sci 136:135–157. doi:10.1016/S0020-0255(01)00147-5 MATHCrossRefGoogle Scholar
  9. Casillas J, Cordon O, Herrera F, Magdalena L (eds) (2003a) Accuracy improvements in linguistic fuzzy modeling. Springer, BerlinGoogle Scholar
  10. Casillas J, Cordon O, Herrera F, Magdalena L (eds) (2003b) Interpretability issues in fuzzy modeling. Springer, BerlinGoogle Scholar
  11. Casillas J, Cordon O, Del Jesus MJ, Herrera F (2005) Genetic tuning of fuzzy rule deep structures preserving interpretability and its interaction with fuzzy rule set reduction. IEEE Trans Fuzzy Syst 13(1):13–29. doi:10.1109/TFUZZ.2004.839670 CrossRefGoogle Scholar
  12. Casillas J, Herrera F, Péreza R, Del Jesus MJ, Villar P (2007) Special issue on genetic fuzzy systems and the Interpretability-Accuracy Trade-off. Int J Approx Reason 44(1):1–3. doi:10.1016/j.ijar.2006.06.002 CrossRefGoogle Scholar
  13. Chang X, Lilly JH (2004) Evolutionary design of a fuzzy classifier from data. IEEE Trans Syst Man Cybern B 34(4):1894–1906. doi:10.1109/TSMCB.2004.831160 CrossRefGoogle Scholar
  14. Chawla N, Bowyer K, Hall L, Kegelmeyer W (2002) Smote: synthetic minority over-sampling technique. J Artif Intell Res 16:321–357MATHGoogle Scholar
  15. Chi Z, Yan H, Pham T (1996) Fuzzy algorithms with applications to image processing and pattern recognition. World Scientific, SingaporeGoogle Scholar
  16. Cococcioni M, Ducange P, Lazzerini B, Marcelloni F (2007) A Pareto-based multi-objective evolutionary approach to the identification of Mamdani fuzzy systems. Soft Comput 11(11):1013–1031. doi:10.1007/s00500-007-0150-6 CrossRefGoogle Scholar
  17. Coello Coello CA, Lamont GB (2004) Applications of multi-objective evolutionary algorithms. World Scientific, SingaporeGoogle Scholar
  18. Coello Coello CA (2006) Evolutionary multi-objective optimization: a historical view of the field. IEEE Comput Intell Mag 1(1):28–36. doi:10.1109/MCI.2006.1597059 CrossRefMathSciNetGoogle Scholar
  19. Cordon O, Del Jesus MJ, Herrera F (1999) A proposal on reasoning methods in fuzzy rule-based classification systems. Int J Approx Reason 20(1):21–45Google Scholar
  20. Cordon O, Herrera F, Hoffmann F, Magdalena L (2001) Genetic fuzzy systems. World Scientific, SingaporeGoogle Scholar
  21. Cordon O, Del Jesus MJ, Herrera F, Magdalena L, Villar P (2003) A multiobjective genetic learning process for joint feature selection and granularity and contexts learning in fuzzy rule-based classification systems. In: Casillas J, Cordon O, Herrera F, Magdalena L (eds) Accuracy improvements in linguistic fuzzy modeling. Springer, Berlin, pp 79–99Google Scholar
  22. Deb K (2001) Multi-objective optimization using evolutionary algorithms. Wiley, LondonGoogle Scholar
  23. Deb K, Pratab A, Agarwal S, Meyarivan T (2002) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput 6(2):182–197. doi:10.1109/4235.996017 CrossRefGoogle Scholar
  24. Duin RPW (2007) PRTools (Version 4.0) A Matlab toolbox for pattern recognition. Pattern recognition group, Delft, University of Technology. http://www.prtools.org
  25. Everson RM, Fieldsend E (2006) Multiobjective optimization of safety related systems: An application short-term conflict alert. IEEE Trans Evol Comput 10(2):187–198. doi:10.1109/TEVC.2005.856067 CrossRefMathSciNetGoogle Scholar
  26. Fawcett T (2003) ROC graphs: Notes and practical considerations for researchers. Tech. Rep. HPL-2003-4, HP LabsGoogle Scholar
  27. Fawcett T (2006) An introduction to ROC analysis. Pattern Recognit Lett 27:861–874. doi:10.1016/j.patrec.2005.10.010 CrossRefGoogle Scholar
  28. Fernandez A, García S, Del Jesus MJ, Herrera F (2008) A study of the behaviour of linguistic fuzzy rule based classification systems in the framework of imbalanced data sets. Fuzzy Sets Syst 159(18):2378–2398. doi:10.1016/j.fss.2007.12.023 CrossRefGoogle Scholar
  29. Herrera F (2008) Genetic fuzzy systems: taxonomy, current research trends and prospects. Evol Intell 1:27–46. doi:10.1007/s12065-007-0001-5 CrossRefGoogle Scholar
  30. Ho SY, Chen HM, Ho SJ, Chen TK (2004) Design of accurate classifiers with a compact fuzzy-rule base using an evolutionary scatter partition of feature space. IEEE Trans Syst Man Cybern 34(2):1031–1044. doi:10.1109/TSMCB.2003.819160 CrossRefGoogle Scholar
  31. Horn J, Nafpliotis N, Goldberg DE (1999) A niched Pareto genetic algorithm for multiobjective optimization. In: Proceedings of the first IEEE conference on evolutionary computation, Orlando, Florida, 27–29 June, pp 82–87Google Scholar
  32. Ishibuchi H (2007) Multiobjective genetic fuzzy systems: review and future research directions. In: Proceedings of the 2007 international conference on fuzzy systems, London, 23–26 July, pp 1-6Google Scholar
  33. Ishibuchi H, Nojima Y (2007) Analysis of interpretability-accuracy tradeoff of fuzzy systems by multiobjective fuzzy genetics-based machine learning. Int J Approx Reason 44(1):4–31. doi:10.1016/j.ijar.2006.01.004 MATHCrossRefMathSciNetGoogle Scholar
  34. Ishibuchi H, Yamamoto T (2004) Fuzzy rule selection by multi-objective genetic local search algorithms and rule evaluation measures in data mining. Fuzzy Sets Syst 141(1):59–88. doi:10.1016/S0165-0114(03)00114-3 MATHCrossRefMathSciNetGoogle Scholar
  35. Ishibuchi H, Yamamoto T (2005) Rule weights specification in fuzzy rule-based classification systems. IEEE Trans Fuzzy Syst 13(4):428–435. doi:10.1109/TFUZZ.2004.841738 CrossRefGoogle Scholar
  36. Ishibuchi H, Murata T, Turksen IB (1997) Single-objective and two-objective genetic algorithms for selecting linguistic rules for pattern classification problems. Fuzzy Sets Syst 89(2):135–150. doi:10.1016/S0165-0114(96)00098-X CrossRefGoogle Scholar
  37. Ishibuchi H, Nakashima T, Nii M (2005a) Classification and modeling with linguistic information granules: advanced approaches to linguistic data Mining. Springer, BerlinGoogle Scholar
  38. Ishibuchi H, Nozaki K, Yamamoto N, Tanaka H (2005b) Selecting fuzzy if-then rules for classification problems using genetic algorithms. IEEE Trans Fuzzy Syst 3(3):260–270. doi:10.1109/91.413232 CrossRefGoogle Scholar
  39. Karr CL, Gentry EJ (1993) Fuzzy control of pH using genetic algorithms. IEEE Trans Fuzzy Syst 1:46–53. doi:10.1109/TFUZZ.1993.390283 CrossRefGoogle Scholar
  40. Knowles JD, Corne DW (2002) Approximating the non dominated front using the Pareto archived evolution strategy. Evol Comput 8(2):149–172. doi:10.1162/106365600568167 CrossRefGoogle Scholar
  41. Kupinski M, Anastasio M (1999) Multiobjective genetic optimization of diagnostic classifiers with implications for generating receiver operating characteristic curves. IEEE Trans Med Imaging 18(8):675–685. doi:10.1109/42.796281 CrossRefGoogle Scholar
  42. Mansoori E, Zolghadri M, Katebi S (2007) A weighting function for improving fuzzy classification systems performance. Fuzzy Sets Syst 158(5):583–591. doi:10.1016/j.fss.2006.10.004 CrossRefMathSciNetGoogle Scholar
  43. Nakashima T, Shaefer G, Yokota Y, Ishibuchi H (2007) A weighted fuzzy classifier and its application to image processing tasks. Fuzzy Sets Syst 158:284–294. doi:10.1016/j.fss.2006.10.011 CrossRefGoogle Scholar
  44. Nauck D, Kruse R (1999) Obtaining interpretable fuzzy classification rules from medical data. Artif Intell Med 16(2):149–169. doi:10.1016/S0933-3657(98)00070-0 CrossRefMathSciNetGoogle Scholar
  45. Provost F, Fawcett T (2001) Robust classification for imprecise environments. Mach Learn 42(3):203–231. doi:10.1023/A:1007601015854 MATHCrossRefGoogle Scholar
  46. Quinlan JR (1993) C4.5: Programs for Machine Learning. Morgan Kauffman, San MateoGoogle Scholar
  47. Raskutti B, Kowalczyk A (2004) Extreme rebalancing for SVMs: a case study. SIGKDD Explor 6(1):60–69. doi:10.1145/1007730.1007739 CrossRefGoogle Scholar
  48. Sheskin D (2003) Handbook of parametric and nonparametric statistical procedures. Chapman & Hall/CRC, London/Boca RatonGoogle Scholar
  49. Srinivas N, Deb K (1998) Multi-objective function optimization using non-dominated sorting genetic algorithms. Evol Comput 2:221–248. doi:10.1162/evco.1994.2.3.221 CrossRefGoogle Scholar
  50. Wilcoxon F (1945) Individual comparisons by ranking methods. Biometrics 1:80–83. doi:10.2307/3001968 CrossRefGoogle Scholar
  51. Woods K, Doss C, Bowyer K, Solka J, Priebe J, Kegelmeyer P (1993) Comparative evaluation of pattern recognition techniques for detection of microcalcifications in mammography. Int J Pattern Recognit Artif Intell 7(6):1417–1436. doi:10.1142/S0218001493000698 CrossRefGoogle Scholar
  52. Yen J, Wang L, Gillespie GW (1998) Improving the interpretability of TSK fuzzy models by combining global learning and local learning. IEEE Trans Fuzzy Syst 6(4):530–537. doi:10.1109/91.728447 CrossRefGoogle Scholar
  53. Zitzler E, Thiele L (1999) Multiobjective evolutionary algorithms: a comparative case study and the strength Pareto approach. IEEE Trans Evol Comput 3:257–271. doi:10.1109/4235.797969 CrossRefGoogle Scholar
  54. Zitzler E, Deb K, Thiele L (2000) Comparison of multiobjective evolutionary algorithms: empirical results. Evol Comput 8(2):173–195. doi:10.1162/106365600568202 CrossRefGoogle Scholar
  55. Zitzler E, Laumanns M, Thiele L (2001) SPEA2: Improving the strength Pareto evolutionary algorithm for multiobjective optimization. In: Proceedings of EUROGEN2001 evolutionary methods for design, opt. and control with applications to industrial problems, Athens, pp 95–100Google Scholar

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  • Pietro Ducange
    • 1
  • Beatrice Lazzerini
    • 1
  • Francesco Marcelloni
    • 1
  1. 1.Dipartimento di Ingegneria dell’Informazione, Elettronica, Informatica, TelecomunicazioniUniversity of PisaPisaItaly

Personalised recommendations