{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T16:32:49Z","timestamp":1772641969623,"version":"3.50.1"},"reference-count":50,"publisher":"Emerald","issue":"2","license":[{"start":{"date-parts":[[2025,4,15]],"date-time":"2025-04-15T00:00:00Z","timestamp":1744675200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.emerald.com\/insight\/site-policies"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJICC"],"published-print":{"date-parts":[[2025,5,8]]},"abstract":"<jats:sec><jats:title content-type=\"abstract-subheading\">Purpose<\/jats:title><jats:p>Takagi-Sugeno-Kang (TSK) fuzzy systems are widely used in classification and regression problems. However, when the data size is large and the feature dimensions are high, the optimization performance of the TSK fuzzy system is poor, and its generalization capability is insufficient. Therefore, this paper proposes an algorithm based on mini-batch gradient descent (MBGD) and batch normalization (BN) to train TSK fuzzy classifiers.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Design\/methodology\/approach<\/jats:title><jats:p>This study uses the AdaDerivative (A) optimizer to optimize the learning rate, weighted firing level (W) applies different weight factors to various features to calculate the firing level of each rule, and adaptive rule firing level loss (A) and knowledge distillation loss (KD) are proposed to encourage each rule to have different average firing levels, as well as to ensure that the output of the rule consequent is close to the mean. Thereby, the final algorithm TSK-BN-2AWKD is obtained. To evaluate the performance of TSK-BN-2AWKD, experiments were conducted on 15 real datasets.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Findings<\/jats:title><jats:p>The experimental results show that A, A, W and KD are effective individually, and the algorithm TSK-BN-2AWKD, after integrating them, can effectively improve the classification performance.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Originality\/value<\/jats:title><jats:p>This study uses the AdaDerivative optimizer to train TSK fuzzy classifiers. Additionally, adaptive rule firing level loss is proposed to assign different expectations to each rule individually. Finally, KD is proposed to force the outputs of the rule consequent to be close to the mean, solving the narrow vision problem that may exist in TSK fuzzy systems.<\/jats:p><\/jats:sec>","DOI":"10.1108\/ijicc-02-2025-0069","type":"journal-article","created":{"date-parts":[[2025,4,14]],"date-time":"2025-04-14T14:15:06Z","timestamp":1744640106000},"page":"397-417","source":"Crossref","is-referenced-by-count":1,"title":["TSK-BN-2AWKD: a TSK fuzzy system optimization method for classification problems"],"prefix":"10.1108","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-7756-148X","authenticated-orcid":false,"given":"Haoting","family":"Chen","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2329-2144","authenticated-orcid":false,"given":"Jialiang","family":"Xie","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-5663-2421","authenticated-orcid":false,"given":"Xinlei","family":"Han","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"140","published-online":{"date-parts":[[2025,4,15]]},"reference":[{"key":"key2025050804575551200_ref001","doi-asserted-by":"publisher","DOI":"10.1016\/j.ins.2024.121101","article-title":"EEG-based TSK fuzzy graph neural network for driver drowsiness estimation","volume":"679","year":"2024","journal-title":"Information Sciences"},{"issue":"4","key":"key2025050804575551200_ref002","doi-asserted-by":"publisher","first-page":"683","DOI":"10.1109\/tfuzz.2011.2181852","article-title":"An integrated mechanism for feature selection and fuzzy rule extraction for classification","volume":"20","year":"2012","journal-title":"IEEE Transactions on Fuzzy Systems"},{"issue":"2","key":"key2025050804575551200_ref003","doi-asserted-by":"publisher","first-page":"209","DOI":"10.1108\/ijicc-03-2012-0010","article-title":"Gain adaptive sliding mode controller based on interval type-II fuzzy neural network designed for attitude control for micro aircraft vehicle","volume":"7","year":"2014","journal-title":"International Journal of Intelligent Computing and Cybernetics"},{"issue":"2","key":"key2025050804575551200_ref004","doi-asserted-by":"publisher","first-page":"1613","DOI":"10.32604\/cmes.2023.027708","article-title":"Cross-domain TSK fuzzy system based on semi-supervised learning for epilepsy classification","volume":"137","year":"2023","journal-title":"Computer Modeling in Engineering and Sciences"},{"issue":"2","key":"key2025050804575551200_ref005","doi-asserted-by":"publisher","first-page":"734","DOI":"10.1109\/tfuzz.2017.2688358","article-title":"Feature selection with controlled redundancy in a fuzzy rule based framework","volume":"26","year":"2018","journal-title":"IEEE Transactions on Fuzzy Systems"},{"key":"key2025050804575551200_ref006","article-title":"Repeated incremental pruning to produce error reduction","year":"1995"},{"key":"key2025050804575551200_ref007","volume-title":"Genetic Fuzzy Systems: Evolutionary Tuning and Learning of Fuzzy Knowledge Bases","year":"2001"},{"issue":"12","key":"key2025050804575551200_ref008","doi-asserted-by":"publisher","first-page":"3065","DOI":"10.1109\/tfuzz.2020.2967282","article-title":"Optimize TSK fuzzy systems for classification problems: mini-batch gradient descent with uniform regularization and batch normalization","volume":"28","year":"2020","journal-title":"IEEE Transactions on Fuzzy Systems"},{"key":"key2025050804575551200_ref009","first-page":"1","article-title":"Curse of dimensionality for tsk fuzzy neural networks: explanation and solutions","year":"2021"},{"issue":"1","key":"key2025050804575551200_ref010","doi-asserted-by":"publisher","first-page":"254","DOI":"10.1109\/tfuzz.2022.3185464","article-title":"Layer normalization for TSK fuzzy system optimization in regression problems","volume":"31","year":"2023","journal-title":"IEEE Transactions on Fuzzy Systems"},{"key":"key2025050804575551200_ref011","first-page":"1","article-title":"Statistical comparisons of classifiers over multiple data sets","volume":"7","year":"2006","journal-title":"Journal of Machine Learning Research"},{"issue":"2","key":"key2025050804575551200_ref012","doi-asserted-by":"publisher","first-page":"210","DOI":"10.1109\/tfuzz.2010.2091961","article-title":"Scalable TSK fuzzy modeling for very large datasets using minimal-enclosing-ball approximation","volume":"19","year":"2011","journal-title":"IEEE Transactions on Fuzzy Systems"},{"issue":"4","key":"key2025050804575551200_ref013","doi-asserted-by":"publisher","first-page":"382","DOI":"10.1108\/ijicc-03-2014-0012","article-title":"Discrete-time indirect adaptive fuzzy control for robot manipulators","volume":"7","year":"2014","journal-title":"International Journal of Intelligent Computing and Cybernetics"},{"key":"key2025050804575551200_ref014","article-title":"Generating accurate rule sets without global optimization","year":"1998"},{"issue":"200","key":"key2025050804575551200_ref015","doi-asserted-by":"publisher","first-page":"675","DOI":"10.2307\/2279372","article-title":"The use of ranks to avoid the assumption of normality implicit in the analysis of variance","volume":"32","year":"1937","journal-title":"Journal of the American Statistical Association"},{"issue":"8","key":"key2025050804575551200_ref016","doi-asserted-by":"publisher","first-page":"3176","DOI":"10.1109\/tfuzz.2021.3106330","article-title":"A concise TSK fuzzy ensemble classifier integrating dropout and bagging for high-dimensional problems","volume":"30","year":"2022","journal-title":"IEEE Transactions on Fuzzy Systems"},{"key":"key2025050804575551200_ref017","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2024.111498","article-title":"A design of fuzzy rule-based classifier optimized through softmax function and information entropy","volume":"156","year":"2024","journal-title":"Applied Soft Computing"},{"issue":"7","key":"key2025050804575551200_ref018","first-page":"38","article-title":"Distilling the knowledge in a neural network","volume":"14","year":"2015","journal-title":"Computer Science"},{"issue":"1","key":"key2025050804575551200_ref019","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1162\/neco.1991.3.1.79","article-title":"Adaptive mixtures of local experts","volume":"3","year":"1991","journal-title":"Neural Computation"},{"issue":"3","key":"key2025050804575551200_ref020","doi-asserted-by":"publisher","first-page":"665","DOI":"10.1109\/21.256541","article-title":"Anfis: adaptive-network-based fuzzy inference system","volume":"23","year":"1993","journal-title":"IEEE Transactions on Systems, Man, and Cybernetics"},{"key":"key2025050804575551200_ref021","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2024.106599","article-title":"Convergence analysis of sparse TSK fuzzy systems based on spectral Dai-Yuan conjugate gradient and application to high-dimensional feature selection","volume":"179","year":"2024","journal-title":"Neural Networks"},{"issue":"4","key":"key2025050804575551200_ref022","doi-asserted-by":"publisher","first-page":"312","DOI":"10.1016\/j.icte.2020.04.010","article-title":"The effect of batch size on the generalizability of the convolutional neural networks on a histopathology dataset","volume":"6","year":"2020","journal-title":"ICT Express"},{"key":"key2025050804575551200_ref023","article-title":"On large-batch training for deep learning: generalization gap and sharp minima","year":"2016","journal-title":"arXiv preprint arXiv:1609.04836"},{"key":"key2025050804575551200_ref024","article-title":"Adam: a method for stochastic optimization","year":"2014","journal-title":"arXiv preprint arXiv:1412.6980"},{"key":"key2025050804575551200_ref025","doi-asserted-by":"publisher","DOI":"10.1016\/j.fss.2024.108949","article-title":"Construction of fuzzy classification systems by fitness sharing based genetic search and boosting based ensemble","volume":"484","year":"2024","journal-title":"Fuzzy Sets and Systems"},{"key":"key2025050804575551200_ref026","article-title":"Adaptive gradient methods with dynamic bound of learning rate","year":"2019","journal-title":"arXiv preprint arXiv:1902.09843"},{"key":"key2025050804575551200_ref027","article-title":"Revisiting small batch training for deep neural networks","year":"2018","journal-title":"arXiv preprint arXiv:1804.07612"},{"key":"key2025050804575551200_ref028","article-title":"Incremental learning for sirms fuzzy systems by Adam method","year":"2017"},{"key":"key2025050804575551200_ref029","volume-title":"Uncertain Rule-Based Fuzzy Systems: Introduction and New Directions","year":"2017","edition":"2nd ed."},{"key":"key2025050804575551200_ref030","article-title":"Distribution-free multiple comparisons","year":"1963"},{"issue":"1","key":"key2025050804575551200_ref031","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1109\/mci.2018.2881644","article-title":"Fuzzy control systems: past, present and future","volume":"14","year":"2019","journal-title":"IEEE Computational Intelligence Magazine"},{"issue":"6","key":"key2025050804575551200_ref032","doi-asserted-by":"publisher","first-page":"1626","DOI":"10.1109\/tsmcb.2008.2006367","article-title":"Simultaneous structure identification and fuzzy rule generation for Takagi\u2013Sugeno models","volume":"38","year":"2008","journal-title":"IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)"},{"key":"key2025050804575551200_ref033","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2024.112010","article-title":"Three-way decision-based Takagi\u2013Sugeno\u2013Kang fuzzy classifier for partially labeled data","volume":"164","year":"2024","journal-title":"Applied Soft Computing"},{"issue":"3","key":"key2025050804575551200_ref034","doi-asserted-by":"publisher","first-page":"400","DOI":"10.1214\/aoms\/1177729586","article-title":"A stochastic approximation method","volume":"22","year":"1951","journal-title":"The Annals of Mathematical Statistics"},{"issue":"2","key":"key2025050804575551200_ref035","doi-asserted-by":"publisher","first-page":"538","DOI":"10.1080\/00051144.2024.2312309","article-title":"Stabilization control of rotary inverted pendulum using a novel EKF-based fuzzy adaptive sliding-mode controller: design and experimental validation","volume":"65","year":"2024","journal-title":"Automatika"},{"key":"key2025050804575551200_ref036","article-title":"Outrageously large neural networks: the sparsely-gated mixture-of-experts layer","year":"2017","journal-title":"arXiv preprint arXiv:1701.06538"},{"key":"key2025050804575551200_ref037","first-page":"5719","article-title":"Mixture models for diverse machine translation: tricks of the trade","year":"2019"},{"key":"key2025050804575551200_ref038","doi-asserted-by":"publisher","first-page":"490","DOI":"10.1016\/j.ins.2021.05.084","article-title":"FCM-RDpA: TSK fuzzy regression model construction using fuzzy c-means clustering, regularization, Droprule, and Powerball AdaBelief","volume":"574","year":"2021","journal-title":"Information Sciences"},{"issue":"2","key":"key2025050804575551200_ref039","doi-asserted-by":"publisher","first-page":"1569","DOI":"10.1109\/tetci.2024.3432667","article-title":"TSK fuzzy system optimization for high-dimensional regression problems","volume":"9","year":"2024","journal-title":"IEEE Transactions on Emerging Topics in Computational Intelligence"},{"issue":"1","key":"key2025050804575551200_ref040","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1109\/tsmc.1985.6313399","article-title":"Fuzzy identification of systems and its applications to modeling and control","volume":"15","year":"1985","journal-title":"IEEE Transactions on Systems, Man, and Cybernetics"},{"key":"key2025050804575551200_ref041","first-page":"1409","article-title":"Back-propagation fuzzy system as nonlinear dynamic system identifiers","year":"1992"},{"key":"key2025050804575551200_ref042","doi-asserted-by":"publisher","first-page":"251","DOI":"10.1016\/j.asoc.2018.12.019","article-title":"Robust fusion algorithm based on RBF neural network with TS fuzzy model and its application to infrared flame detection problem","volume":"76","year":"2019","journal-title":"Applied Soft Computing"},{"issue":"10","key":"key2025050804575551200_ref043","doi-asserted-by":"publisher","first-page":"2570","DOI":"10.1109\/tfuzz.2019.2941697","article-title":"On the functional equivalence of TSK fuzzy systems to neural networks, mixture of experts, CART, and stacking ensemble regression","volume":"28","year":"2020","journal-title":"IEEE Transactions on Fuzzy Systems"},{"issue":"5","key":"key2025050804575551200_ref044","doi-asserted-by":"publisher","first-page":"1003","DOI":"10.1109\/tfuzz.2019.2958559","article-title":"Optimize TSK fuzzy systems for regression problems: mini-batch gradient descent with regularization, droprule, and adabound (MBGD-RDA)","volume":"28","year":"2020","journal-title":"IEEE Transactions on Fuzzy Systems"},{"key":"key2025050804575551200_ref045","doi-asserted-by":"publisher","first-page":"16067","DOI":"10.1609\/aaai.v38i14.29539","article-title":"MoDE: a mixture-of-experts model with mutual distillation among the experts","year":"2024"},{"issue":"11","key":"key2025050804575551200_ref046","doi-asserted-by":"publisher","first-page":"2176","DOI":"10.1109\/tfuzz.2019.2895572","article-title":"Concise fuzzy system modeling integrating soft subspace clustering and sparse learning","volume":"27","year":"2019","journal-title":"IEEE Transactions on Fuzzy Systems"},{"issue":"3","key":"key2025050804575551200_ref047","doi-asserted-by":"publisher","first-page":"601","DOI":"10.1109\/lcsys.2019.2913770","article-title":"On the Powerball method: variants of descent methods for accelerated optimization","volume":"3","year":"2019","journal-title":"IEEE Control Systems Letters"},{"key":"key2025050804575551200_ref048","article-title":"PBSGD: powered stochastic gradient descent methods for accelerated non-convex optimization","year":"2020"},{"key":"key2025050804575551200_ref049","article-title":"Adabelief optimizer: adapting stepsizes by the belief in observed gradients","year":"2020"},{"key":"key2025050804575551200_ref050","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2022.105755","article-title":"Adaderivative optimizer: adapting step-sizes by the derivative term in past gradient information","volume":"119","year":"2023","journal-title":"Engineering Applications of Artificial Intelligence"}],"container-title":["International Journal of Intelligent Computing and Cybernetics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJICC-02-2025-0069\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJICC-02-2025-0069\/full\/html","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T22:54:08Z","timestamp":1753397648000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.emerald.com\/ijicc\/article\/18\/2\/397-417\/1263519"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,15]]},"references-count":50,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,4,15]]},"published-print":{"date-parts":[[2025,5,8]]}},"alternative-id":["10.1108\/IJICC-02-2025-0069"],"URL":"https:\/\/doi.org\/10.1108\/ijicc-02-2025-0069","relation":{},"ISSN":["1756-378X","1756-3798"],"issn-type":[{"value":"1756-378X","type":"print"},{"value":"1756-3798","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4,15]]}}}