Fuzzy rule-based classification system using multi-population quantum evolutionary algorithm with contradictory rule reconstruction
- 99 Downloads
Fuzzy rule-based classification systems (FRBCSs) are appropriate tools for dealing with classification problems because of their interpretable models based on linguistic variables. Intelligent optimization techniques are widely used in the rule mining of FRBCSs due to their parallel processing capability for solving the combination optimization of fuzzy antecedent parameters and “don’t care” variables, however, FRBCSs suffer from misclassification because of the chain coding scheme and non-guidance of updating strategy in rule mining process. This study presents an FRBCS that uses a multi-population quantum evolutionary algorithm with contradictory rule reconstruction. The developed method utilizes fuzzy C-means clustering to heuristically generate representative initial rules, and performs multi-population quantum coding and guided updating to optimize fuzzy rules. Furthermore, contradictory rule reconstruction is introduced to adjust misclassification rules. Numerical experiment results show that the classification accuracy and noise tolerance of the proposed method are better than those of compared FRBCSs. The proposed method is verified by applying it to fault identification in a wind turbine.
KeywordsFuzzy rule-based classification system Multi-population quantum coding Hybrid updating strategy Contradictory rule reconstruction Fault identification
This research is supported by National Natural Science Foundation of China (61102124), Natural Science Foundation of Liaoning Province (20180551032) and Educational Commission of Liaoning Province (LQGD2017035).
- 9.Reddy GT, Khare N (2016) An efficient system for heart disease prediction using hybrid OFBAT with rule-based fuzzy logic model. J Circ Syst Comput 26:379–388Google Scholar
- 12.Sanz JA, Bernardo D, Herrera F, Bustince H, Hagras H (2015) A compact evolutionary interval-valued fuzzy rule-based classification system for the modeling and prediction of real-world financial applications with imbalanced data. Chem Geol 90:973–990Google Scholar
- 14.Tzima FA, Allamanis M, Filotheou A, Mitkas PA (2015) Inducing generalized multi-label rules with learning classifier systems. Comput Sci 12:1–16Google Scholar
- 17.Cheruku R, Edla DR, Kuppili V, Dharavath R (2017) RST-BatMiner: a fuzzy rule miner integrating rough set feature selection and bat optimization for detection of diabetes disease. Appl Soft Comput 6:1–50Google Scholar
- 21.Sharma P, Ratnoo S (2014) A review on discovery of classification rules using pittsburgh approach. Int J Artif Intell Knowl Discov 4:6–16Google Scholar
- 27.Nojima Y, Watanabe K, Ishibuchi H (2015) Effects of heuristic rule generation from multiple patterns in multi-objective fuzzy genetics-based machine learning. In: Proceedings of 2015 ieee congress on evolutionary computation. IEEE, SendaiGoogle Scholar
- 32.Van DBJ, Kaymak U (2002) Fuzzy classification using probability-based rule weighting. In: Proceedings of 2002 IEEE international conference on fuzzy systems. IEEE, HonoluluGoogle Scholar
- 33.Zhang YX, Qian XY, Peng HD, Wang JH (2016) An allele real-coded quantum evolutionary algorithm based on hybrid updating strategy. Comput Intell Neurosci 9:50–60Google Scholar
- 34.Bache K, Lichman M (2013) UCI machine learning repository. http://archive.ics.uci.edu/ml/datasets.html
- 41.Jonkman J, Butterfield S, Musial W (2009) Definition of a 5-MW reference wind turbine for offshore system development. Office of Scientific and Technical Information Technical Reports 2:1–75Google Scholar
- 42.Odgaard PF, Johnson KE (2013) Wind turbine fault detection and fault tolerant control: an enhanced benchmark challenge. In: Proceedings of IEEE 2013 American control conference. IEEE, Washington, pp 4447–4452Google Scholar