Abstract
When a flexible fuzzy rule structure such as those with antecedent in conjunctive normal form is used, the interpretability of the obtained fuzzy model is significantly improved. However, some important problems appear related to the interaction among this set of rules. Indeed, it is relatively easy to get inconsistencies, lack of completeness, redundancies, etc. Generally, these properties are ignored or mildly faced. This paper, however, focuses on the design of a multiobjective genetic algorithm that properly considers all these properties thus ensuring an effective search space exploration and generation of highly legible and accurate fuzzy models.
Similar content being viewed by others
Notes
L. Torgo. Collection of regression datasets. http://www.liacc.up.pt/~ltorgo/Regression/DataSets.html.
J. Casillas. FMLib: fuzzy modeling library. http://decsai.ugr.es/~casillas/FMLib/.
KEEL: Knowledge extraction based on evolutionary learning. http://www.keel.es.
References
Carmona P, Castro J (2005) Interpretability enhancement of fuzzy modeling using ant colony optimization. In: Proceedings of the 1st international workshop on genetic fuzzy systems (GFS 2005), Granada, Spain, pp 148–153
Carmona P, Castro J, Zurita J (2004) Learning maximal structure fuzzy rules with exceptions. Fuzzy Sets Syst 146:63–77
Casillas J, Martínez-López F (2008) Mining uncertain data with multiobjective genetic fuzzy systems to be applied in consumer behaviour modelling. Expert Syst Appl (in press). doi:10.1016/j.eswa.2007.11.035
Casillas J, Cordón O, Herrera F, Magdalena L (eds) (2003a) Accuracy improvements in linguistic fuzzy modeling. Springer, Heidelberg
Casillas J, Cordón O, Herrera F, Magdalena L (eds) (2003b) Interpretability issues in fuzzy modeling. Springer, Heidelberg
Casillas J, Cordón O, del Jesus M, Herrera F (2005a) 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
Casillas J, Cordón O, de Viana IF, Herrera F (2005b) Learning cooperative linguistic fuzzy rules using the best-worst ant system algorithm. Int J Intell Syst 20:433–452
Castro J, Castro-Schez J, Zurita J (1999) Learning maximal structure rules in fuzzy logic for knowledge acquisition in expert systems. Fuzzy Sets Syst 101(3):331–342
Cordón O, Herrera F, Sánchez L (1999) Solving electrical distribution problems using hybrid evolutionary data analysis techniques. Appl Intell 10(1):5–24
Cordón O, Herrera F, Hoffmann F, Magdalena L (2001) Genetic fuzzy systems: evolutionary tuning and learning of fuzzy knowledge bases. World Scientific, Singapore
Deb K, Pratap A, Agarwal S, Meyarevian T (2002) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evolut Comput 6(2):182–197
Fullér R (2000) Introduction to neuro-fuzzy systems. Physica-Verlag, Heidelberg
González A, Pérez R (1998) Completeness and consistency conditions for learning fuzzy rules. Fuzzy Sets Syst 96(1):37–51
González A, Pérez R (1999) SLAVE: a genetic learning system based on an iterative approach. IEEE Trans Fuzzy Syst 7(2):176–191
Hastie T, Tibshirani R (1990) Generalized additive models. Chapman & Hall, London
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
Ishibuchi H, Yamamoto T, Nakashima T (2006) An approach to fuzzy default reasoning for function approximation. Soft Comput 10:850–864
Jin Y, von Seelen W, Sendhoff B (1999) On generating FC3 fuzzy rule systems from data using evolution strategies. IEEE Trans Syst Man Cybern B Cybern 29(4):829–845
Karr C (1991) Genetic algorithms for fuzzy controllers. AI Expert 6(2):26–33
Lavrač N, Cestnik B, Gamberger D, Flach P (2004) Decision support through subgroup discovery: three case studies and the lessons learned. Mach Learn 57(1–2):115–143
Magdalena L (1997) Adapting the gain of an FLC with genetic algorithms. Int J Approx Reason 17(4):327–349
Nauck D, Klawonn F, Kruse R (1997) Foundations of neuro-fuzzy systems. Wiley, New York
Nozaki K, Ishibuchi H, Tanaka H (1997) A simple but powerful heuristic method for generating fuzzy rules from numerical data. Fuzzy Sets Syst 86(3):251–270
Otero J, Sánchez L (2006) Induction of descriptive fuzzy classifiers with the logitboost algorithm. Soft Comput 10(9):825–835
Sánchez L, Couso I (2000) Fuzzy random variables-based modeling with GA-P algorithms. In: In Information, uncertainty and fusion. Kluwer, Dordrecht, pp 245–256
Sánchez L, Couso I, Corrales JA (2001) Combining GP operators with SA search to evolve fuzzy rule based classifiers. Inf Sci 136(1–4):175–191
Thrift P (1991) Fuzzy logic synthesis with genetic algorithms. In: Belew R, Booker L (eds) Proceedings of the 4th international conference on genetic algorithms. Morgan Kaufmann Publishers, San Mateo, pp 509–513
Valenzuela-Rendón M (1991) The fuzzy classifier system: a classifier system for continuously varying variables. In: Proceedings of the 4th international conference on genetic algorithms. Morgan Kaufmann Publishers, San Mateo, pp 346–353
Wang C, Hong T, Tseng S, Liao C (1998) Automatically integrating multiple rules sets in a distributed-knowledge environment. IEEE Trans Syst Man Cybern C Appl Rev 28(3):471–476
Wang H, Kwong S, Jin Y, Wei W, Man KF (2005) Agent-based evolutionary approach for interpretable rule-based knowledge extraction. IEEE Trans Syst Man Cybern 35(2):143–155
Wang LX (2003) The WM method completed: a flexible fuzzy system approach to data mining. IEEE Trans Fuzzy Syst 11(6):768–782
Wang LX, Mendel J (1992) Generating fuzzy rules by learning from examples. IEEE Trans Syst Man Cybern 22(6):1414–1427
Weigend A, Gershenfeld N (eds) (1993) Time series prediction: forecasting the future and understanding the past. In: 1992 NATO Advanced Research Workshop on Comparative Time Series Analysis. Addison-Wesley, Santa Fe
Xiong N, Litz L (2000) Fuzzy modeling based on premise optimization. In: Proceedings of the 9th IEEE international conference on fuzzy systems, San Antonio, TX, USA, pp 859–864
Yager R (1993) On a hierarchical structure for fuzzy modeling and control. IEEE Trans Syst Man Cybern 23(4):1189–1197
Acknowledgments
This work was supported in part by the Spanish Ministry of Education and Science under grant no. TIN2005-08386-C05-01 and the Andalusian Government under grants no. P05-TIC-00531 and P07-TIC-3185.
Author information
Authors and Affiliations
Corresponding author
Appendix: Wang–Mendel algorithm
Appendix: Wang–Mendel algorithm
The ad hoc data-driven Mamdani-type fuzzy rule set generation process proposed by Wang and Mendel(1992) is widely known and used because of its simplicity. In our algorithm, Pitts-DNF, it is used in the initialization process and completeness operator. Therefore, for the sake of readability we briefly introduce the algorithm in this appendix.
It is based on working with an input-output data pair set representing the behavior of the problem being solved:
with N being the data set size, n the number of input variables, and m the number of output variables. The algorithm consists of the following steps:
-
1.
Consider a fuzzy partition (definitions of the membership functions parameters) for each input/output variable.
-
2.
Generate a candidate fuzzy rule set: This set is formed by the rule best covering each example contained in E. Thus, N candidate fuzzy rules, CR l, are obtained. The structure of each rule is generated by taking a specific example, i.e., an (n + m)-dimensional real vector, and setting each one of the variables to the linguistic term (associated fuzzy set) best covering every vector component:
$$ \begin{aligned} CR^l: &\hbox{IF }X_1\hbox{ is }A_1^l\hbox{ and }\ldots\hbox{ and }X_n\hbox{ is }A_n^l\\ &\hbox{THEN }Y_1\hbox{ is }B^l_1\hbox{ and }\ldots\hbox{ and }Y_m\hbox{ is }B^l_m \end{aligned} $$$$ A_i^l = \hbox{arg }\mathop {\max }\limits_{A^{\prime} \in {\mathbf{A}}_i } \mu _{A^{\prime}} (x_i^l ),\quad B^l_j = \hbox{arg } \mathop {\max }\limits_{B^{\prime}\in {\mathbf{B}}_{\mathbf j}} \mu _{B^{\prime}} (y^l_j ) $$ -
3.
Give an importance degree to each candidate rule:
$$ D(CR^l) = \prod_{i=1}^n{\mu_{A^l_i}(x^l_i)} \cdot \prod_{j=1}^{m}{\mu_{B^l_j}(y^l_j)} $$ -
4.
Obtain a final fuzzy rule set from the candidate fuzzy rule set: To do so, the N candidate rules are first grouped in g different groups, each one of them composed of all the candidate rules containing the same antecedent combination. To build the final fuzzy rule set, the rule with the highest importance degree is chosen in each group. Hence, g will be both the number of different antecedent combinations in the candidate rule set and the number of rules in the Mamdani-type fuzzy rule set finally generated.
Rights and permissions
About this article
Cite this article
Casillas, J., Martínez, P. & Benítez, A.D. Learning consistent, complete and compact sets of fuzzy rules in conjunctive normal form for regression problems. Soft Comput 13, 451–465 (2009). https://doi.org/10.1007/s00500-008-0361-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-008-0361-5