Stacking Hybrid System of Computational Intelligence Based on Kernel Activation-Membership Functions and its Online Learning in Pattern Recognition Task

Authors

  • Oleh Zolotukhin Kharkiv National University of Radio Electronics
  • Yevgeniy Bodyansky Kharkiv National University of Radio Electronics
  • Valentin Filatov Kharkiv National University of Radio Electronics
  • Maryna Kudryavtseva Kharkiv National University of Radio Electronics
  • Oleksandr Vasylets Kharkiv National University of Radio Electronics

DOI:

https://doi.org/10.30837/2522-9818.2026.2.040

Keywords:

computational intelligence; data stream mining; kernel activation-membership function; machine learning; neuro-neo-fuzzy system; stacking hybrid system

Abstract

Subject of research. The subject of the research is a stacking hybrid computational intelligence system based on a kernel activation-membership function, which combines the advantages of neural networks and fuzzy models for solving pattern recognition tasks under conditions of uncertainty, noise, and limited dimensionality of data sets. The goal of the work. The goal of the work is to develop a stacking hybrid system with a kernel activation-membership function and online learning to improve the accuracy, stability, and speed of pattern recognition in non-stationary environments in real-time. Task. Develop the architecture of a stacking hybrid system; propose a kernel activation-membership function and an algorithm for its parameterization; construct an online learning algorithm with step-by-step parameter updates; investigate the properties of convergence and computational complexity; conduct an experimental comparison with basic machine learning models and classical neuro-fuzzy approaches. Methods. Machine learning methods, ensemble and stacking approaches, kernel methods, adaptive optimization in online mode, statistical analysis of classification quality, modelling and computational experiments on synthetic and real datasets applied to pattern recognition tasks. Results. The proposed model provides higher classification accuracy and better generalization ability compared to individual baseline models, demonstrates resistance to noise and changes in data distributions, as well as rapid adaptation in streaming data conditions thanks to online learning. It is shown to reduce the error and ensure stable operation of the system in non-stationary environments. Conclusions. The developed stacking hybrid system with a kernel activation-membership function is an effective tool for real-time pattern recognition tasks. The combination of the stacking approach and online learning increases the accuracy, stability and adaptability of the system, which makes the proposed approach promising for practical application in intelligent information systems and cyber-physical environments.

Downloads

Download data is not yet available.

Author Biographies

Oleh Zolotukhin, Kharkiv National University of Radio Electronics

Candidate of Technical Sciences, Associate Professor, Dean of the Computer Science Faculty

Yevgeniy Bodyansky, Kharkiv National University of Radio Electronics

Doctor of Technical Science, Professor, Professor of the Artificial Intelligence Department

Valentin Filatov, Kharkiv National University of Radio Electronics

Doctor of Technical Science, Professor, Professor of the Artificial Intelligence Department

Maryna Kudryavtseva, Kharkiv National University of Radio Electronics

Candidate of Technical Sciences, Associate Professor, Professor of the Artificial Intelligence Department

Oleksandr Vasylets, Kharkiv National University of Radio Electronics

Postgraduate Student of the Artificial Intelligence Department

References

References

Mumford, C.L. and Jain, L.C. (2009), Computational Intelligence: Collaboration, Fusion and Emergence, Springer, Berlin Heidelberg, 732 p. DOI: https://doi.org/10.1007/978-3-642-01799-5

Kruse, R., Borgelt, C., Braune, C., Mostaghim, S. and Steinbrecher, M. (2016), Computational Intelligence: A Methodological Introduction, Springer, London, 564 p. DOI: https://doi.org/10.1007/978-1-4471-7296-3

Kacprzyk, J. and Pedrycz, W. (eds.) (2015), Springer Handbook of Computational Intelligence, Springer, Berlin Heidelberg, 1634 p. DOI: https://doi.org/10.1007/978-3-662-43505-2

Goodfellow, I., Bengio, Y. and Courville, A. (2016), Deep Learning, MIT Press, Cambridge, MA, 800 p. ISBN 978-0-262-03561-3.

Schmidhuber, J. (2015), "Deep Learning in Neural Networks: An Overview", Neural Networks, Vol. 61, pp. 85–117. DOI: https://doi.org/10.1016/j.neunet.2014.09.003

Kung, S.-Y. (2014), Kernel Methods and Machine Learning, Cambridge University Press, Cambridge, 591 p. DOI: https://doi.org/10.1017/CBO9781139176224

Karamichailidou, D., Gerolymatos, G., Patrinos, P., Sarimveis, H. and Alexandridis, A. (2024), "Radial Basis Function Neural Network Training Using Variable Projection and Fuzzy Means", Neural Computing and Applications, Vol. 36, No. 33, pp. 21137–21151. DOI: https://doi.org/10.1007/s00521-024-10274-3

Ismayilova, A. and Ismayilov, V.E. (2024), "On the Universal Approximation Property of Radial Basis Function Neural Networks", Annals of Mathematics and Artificial Intelligence, Vol. 92, No. 3, pp. 174–185. DOI: https://doi.org/10.1007/s10472-023-09901-x

Marrero, I. (2025), "Relaxed Conditions for Universal Approximation by Radial Basis Function Neural Networks of Hankel Translates", AIMS Mathematics, Vol. 10, No. 5, pp. 10852–10865. DOI: https://doi.org/10.3934/math.2025493

Jensen, V., Bianchi, F.M. and Anfinsen, S.N. (2024), "Ensemble Conformalized Quantile Regression for Probabilistic Time Series Forecasting", IEEE Transactions on Neural Networks and Learning Systems, Vol. 35, No. 7, pp. 9014–9025. DOI: https://doi.org/10.1109/TNNLS.2022.3217694

Zolotukhin, O.V., Kudryavtseva, M.S., Bodyanskiy, Y.V., Filatov, V.O., Antilikatorov, A.V. and Kalinin, D.V. (2026), "Physical-Informed Neural Network in Signal Processing and Network Traffic Communications", System Research and Information Technologies, No. 1, pp. 155–169. DOI: https://doi.org/10.20535/SRIT.2308-8893.2026.1.11

Angelov, P., Filev, D. and Kasabov, N. (eds.) (2010), Evolving Intelligent Systems: Methodology and Applications, Wiley/IEEE Press, Hoboken, NJ, 444 p. DOI: https://doi.org/10.1002/9780470569962

Angelov, P., Zhou, X. and Lughofer, E. (2008), "Evolving Fuzzy Classifiers Using Different Model Architectures", Fuzzy Sets and Systems, Vol. 159, No. 23, pp. 3160–3182. DOI: https://doi.org/10.1016/j.fss.2008.06.019

Baruah, R.D., Angelov, P. and Baruah, D. (2020), "Evolving Intelligent Systems", In: Wiley Encyclopedia of Electrical and Electronics Engineering, pp. 1–17. DOI: https://doi.org/10.1002/047134608X.W8405

Filatov, V., Zolotukhin, O. and Kudryavtseva, M. (2025), "Intellectual Data Analysis in Relational Information and Analytical Systems", Innovative Technologies and Scientific Solutions for Industries, Vol. 4, No. 34, pp. 101–111. DOI: https://doi.org/10.30837/2522-9818.2025.4.101

Jiang, W., Chen, Z., Xiang, Y., Shao, D., Ma, L. and Zhang, J. (2019), "SSEM: A Novel Self-Adaptive Stacking Ensemble Model for Classification", IEEE Access, Vol. 7, pp. 120337–120349. DOI: https://doi.org/10.1109/ACCESS.2019.2933262

Uchino, E. and Yamakawa, T. (1994), "Neo-Fuzzy-Neuron Based New Approach to System Modeling, with Application to Actual System", In: Proceedings of the Sixth International Conference on Tools with Artificial Intelligence (TAI'94), pp. 701–706. DOI: https://doi.org/10.1109/TAI.1994.346442

Uchino, E. and Yamakawa, T. (1997), "Soft Computing Based Signal Prediction, Restoration and Filtering", In: Intelligent Hybrid Systems: Fuzzy Logic, Neural Networks and Genetic Algorithms, Vol. 413, pp. 331–351. DOI: https://doi.org/10.1007/978-1-4615-6191-0_14

Brereton, R.G. and Lloyd, G.R. (2010), "Support Vector Machines for Classification and Regression", Analyst, Vol. 135, pp. 230–267. DOI: https://doi.org/10.1039/B918972F

Aggarwal, C.C. (2018), Neural Networks and Deep Learning: A Textbook, Springer, Cham, 520 p. DOI: https://doi.org/10.1007/978-3-319-94463-0

Filatov, V., Semenets, V. and Zolotukhin, O. (2019), "Synthesis of Semantic Model of Subject Area at Integration of Relational Databases", In: Proceedings of the International Conference on Advanced Optoelectronics and Lasers (CAOL), pp. 598–601. DOI: https://doi.org/10.1109/CAOL46282.2019.9019532

Bodyanskiy, Y., Zolotukhin, O., Yerokhin, A., Kudryavtseva, M. and Yerokhin, M. (2025), "Fast Stacking Neuro-Neo-Fuzzy System for Inverse Modeling in Online Mode", International Journal of Computing, Vol. 24, No. 4, pp. 661–667. DOI: https://doi.org/10.47839/ijc.24.4.4330

Lemos, A., Caminhas, W.M. and Gomide, F. (2014), "A Fast Learning Algorithm for Evolving Neo-Fuzzy Neuron", Applied Soft Computing, Vol. 14, Part B, pp. 194–209. DOI: https://doi.org/10.1016/j.asoc.2013.03.022

Zurita, D., Delgado, M., Carino, J.A., Ortega, J.A. and Clerc, G. (2016), "Industrial Time Series Modelling by Means of the Neo-Fuzzy Neuron", IEEE Access, Vol. 4, pp. 6151–6160. DOI: https://doi.org/10.1109/ACCESS.2016.2611649

Chang, W.J., Chang, C.-H. and Ku, C.C. (2010), "Fuzzy Controller Design for Takagi–Sugeno Fuzzy Models with Multiplicative Noises via Relaxed Non-Quadratic Stability Analysis", Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, Vol. 224, No. 8, pp. 918–931. DOI: https://doi.org/10.1243/09596518JSCE1035

Mokodompit, M., Nasib, S.K., Djakaria, I., Yahya, N.I. and Hasan, I.K. (2025), "Implementation of Fuzzy Time Series Markov Chain Method Using Kernel Smoothing in Forecasting the Stock Price of PT. Elnusa Tbk.", Indonesian Journal of Computational and Applied Mathematics, Vol. 1, No. 1, pp. 18–28. DOI: https://doi.org/10.64182/indocam.v1i1.9

Chen, Y.-C., Genovese, C.R. and Wasserman, L. (2016), "A Comprehensive Approach to Mode Clustering", Electronic Journal of Statistics, Vol. 10, No. 1, pp. 210–241. https://doi.org/10.1214/16-EJS1115

Needell, D. and Tropp, J.A. (2014), "Paved with Good Intentions: Analysis of a Randomized Block Kaczmarz Method", Linear Algebra and its Applications, Vol. 441, pp. 199–221. DOI: https://doi.org/10.1016/j.laa.2013.01.022

Sayed, A.H. (2014), "Adaptive Networks", Proceedings of the IEEE, Vol. 102, No. 4, pp. 460–497. DOI: https://doi.org/10.1109/JPROC.2014.2306253

Buhmann, M.D. (2003), Radial Basis Functions: Theory and Implementations, Cambridge University Press, Cambridge, 272 p. DOI: https://doi.org/10.1017/CBO9780511543241

Kohonen, T. (1995), Self-Organizing Maps, Springer, Berlin Heidelberg, 362 p. DOI: https://doi.org/10.1007/978-3-642-97610-0

Zolotukhin, O., Filatov, V., Yerokhin, A., Kudryavtseva, M. and Semenets, V. (2021), "An Approach to the Selection of Behavior Patterns of Autonomous Intelligent Mobile Systems", In: Proceedings of the IEEE 8th International Conference on Problems of Infocommunications, Science and Technology (PIC S&T), pp. 349–352. DOI: https://doi.org/10.1109/PICST54195.2021.9772110

Zolotukhin, O., Filatov, V., Yerokhin, A., Kudryavtseva, M. and Semenets, V. (2021), "The Methods for the Prediction of Climate Control Indicators in the Internet of Things Systems", CEUR Workshop Proceedings, Vol. 3013, pp. 391–400. DOI: https://doi.org/10.5281/zenodo.14526027

Dashenkov, D. and Smelyakov, K. (2025), "Extending the ImageNET Dataset for Multimodal Text and Image Learning", Innovative Technologies and Scientific Solutions for Industries, Vol. 1, No. 31, pp. 20–31. DOI: https://doi.org/10.30837/2522-9818.2025.1.020

Danylenko, S. and Smelyakov, K. (2025), "Content-Based Image Retrieval Method in a Multidimensional Data Model at Big Data Scale", Innovative Technologies and Scientific Solutions for Industries, Vol. 4, No. 34, pp. 18–31. DOI: https://doi.org/10.30837/2522-9818.2025.4.018

Amirian, M. and Schwenker, F. (2020), "Radial Basis Function Networks for Convolutional Neural Networks to Learn Similarity Distance Metric and Improve Interpretability", IEEE Access, Vol. 8, pp. 123087–123097. DOI: https://doi.org/10.1109/ACCESS.2020.3007337

Downloads

Published

2026-06-27

How to Cite

Zolotukhin, O., Bodyansky, Y., Filatov, V., Kudryavtseva, M. and Vasylets, O. (2026) “Stacking Hybrid System of Computational Intelligence Based on Kernel Activation-Membership Functions and its Online Learning in Pattern Recognition Task”, INNOVATIVE TECHNOLOGIES AND SCIENTIFIC SOLUTIONS FOR INDUSTRIES, (2(36), pp. 40–51. doi: 10.30837/2522-9818.2026.2.040.

Issue

No. 2(36) (2026): Innovative Technologies and Scientific Solutions for Industries

Section

INFORMATION TECHNOLOGY

License

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Our journal abides by the Creative Commons copyright rights and permissions for open access journals.

Authors who publish with this journal agree to the following terms:

  • Authors hold the copyright without restrictions and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0) that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.

  • Authors are able to enter into separate, additional contractual arrangements for the non-commercial and non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.

  • Authors are permitted and encouraged to post their published work online (e.g., in institutional repositories or on their website) as it can lead to productive exchanges, as well as earlier and greater citation of published work.