×

A proposed method for learning rule weights in fuzzy rule-based classification systems. (English) Zbl 1176.68161

Summary: In Fuzzy Rule-Based Classification Systems (FRBCSs), rule weighting has often been used as a simple mechanism to tune the classifier. In past research, a number of heuristic rule weight specification methods have been proposed for this purpose. A learning algorithm based on reward and punishment has also been proposed to adjust the weights of each fuzzy rule in the rule-base. In this paper, a new method of learning rule weight in FRBCSs is proposed. The method can be used when single winner or weighted vote methods of reasoning is used. Compared with reward and punishment scheme, the proposed method is much faster and more effective. Another advantage of the proposed method is that, during the learning process, redundant rules are removed (i.e., by setting their weights to zero). The final rule-base usually contains much fewer rules than the initial one. This feature is very useful since a compact rule-base is usually desired from the point of efficiency and interpretability. A number of UCI data sets are used to assess the performance of the proposed method in comparison with reward and punishment scheme.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Abe, S.; Lan, M. S., A method for fuzzy rules extraction directly from numerical data and its application to pattern classification, IEEE Trans. Fuzzy Systems, 3, 1, 18-28 (1995)
[2] Abe, S.; Thawonmas, R., A fuzzy classifier with ellipsoidal regions, IEEE Trans. Fuzzy Systems, 5, 3, 358-368 (1997)
[3] Agrawal, R.; Srikant, R., Fast algorithms for mining association rules, (Proc. 20th Internat. Conf. on Very Large Databases (1994)), 487-499
[4] Casillas, J.; Cordón, O.; Del Jesus, M. J.; Herrera, F., Genetic feature selection in a fuzzy rule-based classification system learning process for high-dimensional problems, Inform. Sci., 136, 1-4, 135-157 (2001) · Zbl 0996.68158
[5] Chung Hu, Y.; Hshiung Tzeng, G., Elicitation of classification rules by fuzzy data mining, Engrg. Appl. of Artificial Intelligence, 16, 7-8, 709-716 (2003)
[6] Cordon, O.; del Jesus, M. J.; Herrera, F., A proposal on reasoning methods in fuzzy rule-based classification systems, Internat. J. Approx. Reason., 20, 21-45 (1999)
[7] De Cock, M.; Cornelis, C.; Kerre, E. E., Elicitation of fuzzy association rules from positive and negative examples, Fuzzy Sets and Systems, 149, 1, 73-85 (2005) · Zbl 1085.68169
[8] T. Fawcett, ROC graphs: notes and practical considerations for researchers, Technical Report HPL-2003-4, HP Labs, 2003.; T. Fawcett, ROC graphs: notes and practical considerations for researchers, Technical Report HPL-2003-4, HP Labs, 2003.
[9] Gómez-Skarmeta, A. F.; Valdés, M.; Jiménez, F.; Marín-Blázquez, J. G., Approximative fuzzy rules approaches for classification with hybrid-GA techniques, Inform. Sci., 136, 1-4, 193-214 (2001) · Zbl 1002.68778
[10] Gonzalez, A.; Perez, R., SLAVE: a genetic learning system based on an iterative approach, IEEE Trans. Fuzzy Systems, 7, 2, 176-191 (1999)
[11] Ishibuchi, H.; Nakashima, T.; Morisawa, T., Voting in fuzzy rule-based systems for pattern classification problems, Fuzzy Sets and Systems, 103, 2, 223-238 (1999)
[12] Ishibuchi, H.; Nakashima, T.; Murata, T., Three-objective genetics-based machine learning for linguistic rule extraction, Inform. Sci., 136, 1-4, 109-133 (2001) · Zbl 0996.68175
[13] Ishibuchi, H.; Nozaki, K.; Tanaka, H., Distributed representation of fuzzy rules and its application to pattern classification, Fuzzy Sets and Systems, 52, 1, 21-32 (1992)
[14] Ishibuchi, H.; Yamamoto, T., Comparison of heuristic criteria for fuzzy rule selection in classification problems, Fuzzy Optimization and Decision Making, 3, 2, 119-139 (2004) · Zbl 1062.68116
[15] Ishibuchi, H.; Yamamoto, T., Fuzzy rule selection by multi-objective genetic local search algorithms and rule evaluation measures in data mining, Fuzzy Sets and Systems, 141, 1, 59-88 (2004) · Zbl 1081.68091
[16] Ishibuchi, H.; Yamamoto, T., Rule weight specification in fuzzy rule-based classification systems, IEEE Trans. Fuzzy Systems, 13, 4, 428-435 (2005)
[17] Lachiche, N.; Flach, P., Improving accuracy and cost of two-class and multi-class probabilistic classifiers using ROC curves, (Proc. 20th Internat. Conf. on Machine Learning (ICML’03) (2003)), 416-423
[18] Mansoori, E. G.; Zolghadri, M. J.; Katebi, S. D., A weighting function for improving fuzzy classification systems performance, Fuzzy Sets and Systems, 158, 5, 583-591 (2007) · Zbl 1161.68800
[19] Mitra, S.; Kuncheva, L. I., Improving classification performance using fuzzy MLP and two-level selective partitioning of the feature space, Fuzzy Sets and Systems, 70, 1, 1-13 (1995)
[20] Nauck, D.; Kruse, R., A neuro-fuzzy method to learn fuzzy classification rules from data, Fuzzy Sets and Systems, 89, 3, 277-288 (1997)
[21] Nauck, D.; Kruse, R., How the learning of rule weights affects the interpretability of fuzzy systems, (Proc. 7th IEEE Internat. Conf. on Fuzzy Systems (1998)), 1235-1240
[22] Nozaki, K.; Ishibuchi, H.; Tanaka, H., Adaptive fuzzy rule-based classification systems, IEEE Trans. Fuzzy Systems, 4, 3, 238-250 (1996)
[23] Roubos, J. A.; Setnes, M.; Abonyi, J., Learning fuzzy classification rules from labeled data, Inform. Sci., 150, 1-2, 77-93 (2003)
[24] Sánchez, L.; Couso, I.; Corrales, J. A.; Cordón, O.; Del Jesus, M. J.; Herrera, F., Combining GP operators with SA search to evolve fuzzy rule based classifiers, Inform. Sci., 136, 1-4, 175-191 (2001) · Zbl 0996.68176
[25] Uebele, V.; Abe, S.; Lan, M. S., A neural-network-based fuzzy classifier, IEEE Trans. Systems Man Cybernet., 25, 2, 353-361 (1995)
[26] Zolghadri, M. J.; Mansoori, E. G., Weighting fuzzy classification rules using receiver operating characteristics (ROC) analysis, Inform. Sci., 177, 11, 2296-2307 (2007)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.