JBT Journal
Advanced

Raqamli Transformatsiya va Sun’iy Intellekt

SUST SHAKLLANGAN JARAYONLARNI BASHORATLASHDA MOSLASHUVCHAN ANSAMBL METODOLOGIYASI

Pulatov G‘iyos Gofurjonovich1

1 “TIQXMMI” Milliy tadqiqot universiteti

Zenodo DOI: 10.5281/zenodo.22668032

Keywords

moslashuvchan bashoratlash kompetentlik bahosi dinamik ansambl seleksiyasi mustahkamlash mexanizmi metamodel

Abstract

Bashoratlash modellarini bir marta hisoblangan og‘irliklar bilan birlashtirish, garchi statik yondashuvlarga nisbatan yutuqqa ega bo‘lsa-da, vaqt o‘tishi bilan ma’lumot xususiyatlari o‘zgarganda o‘z samaradorligini yo‘qota boshlaydi. Ushbu ishda sust shakllangan jarayonlarni bashoratlash uchun uchta bosqichni - vaznli ovoz berish, robast kombinatsiya va prognostik xatoga asoslangan dinamik seleksiyani - ketma-ket birlashtiruvchi moslashuvchan metamodel taklif qilinadi. Metodning o‘ziga xosligi shundaki, ansambl tarkibidagi eng kompetent algoritm har bir vaqt oynasi uchun barcha kombinatsiyalarni sanab chiqmasdan, kelgusi xatoni oldindan bashoratlovchi yordamchi tarmoq orqali tanlanadi, tanlov sifati esa modellarni mukofotlash va jarimalash mexanizmi orqali vaqt davomida sayqallanadi. Yurak-qon tomir kasalliklari bo‘yicha bemorlar murojaatini bashoratlash misolida o‘tkazilgan tajriba shuni ko‘rsatdiki, taklif etilgan yondashuv statik kombinatsiyaga nisbatan ham qo‘shimcha yutuq beradi, biroq bu yutuqning qay darajada dinamik seleksiya hisobiga, qay darajada esa boshqa omillar hisobiga ekanligini aniqlash uchun qo‘shimcha, ayirmachi tadqiqotlar zarur - bu masala maqolaning muhokama qismida ochiq muammo sifatida ko‘tarib chiqilgan.

Views: 112 Published: 2026-08-28 Pages: 185-190 Volume: 4 Issue: 4

References

  1. Cruz R.M.O., Sabourin R., Cavalcanti G.D.C. Dynamic classifier selection: Recent advances and perspectives // Information Fusion. – 2018. – Vol. 41. – P. 195–216.
  2. Moura T.J.M., Cavalcanti G.D.C., Oliveira L.S. MINE: A framework for dynamic regressor selection // Information Sciences. – 2021. – Vol. 543. – P. 157–179.
  3. Zhang Z.-L., Liu X.-W., Yan W.-J. DES-AS: Dynamic ensemble selection based on algorithm Shapley // Pattern Recognition. – 2025. – Vol. 158. – Article 111062.
  4. Pulatov G., Pulatova G. Neural network prediction of the frequency of admission of patients with cardiovascular diseases to clinical institutions // Eureka Journal of Artificial Intelligence and Data Innovation (EJAIDI). – 2026. – Vol. 2, No. 3. – P. 46–54.
  5. Britto A.S. Jr., Sabourin R., Oliveira L.E.S. Dynamic selection of classifiers – A comprehensive review // Pattern Recognition. – 2014. – Vol. 47, No. 11. – P. 3665–3680.
  6. Krogh A., Vedelsby J. Neural network ensembles, cross validation, and active learning // Advances in Neural Information Processing Systems. – 1995. – Vol. 7. – P. 231–238.
  7. Pulatov G. Development of an ANFIS neuro-fuzzy model for forecasting weakly structured processes // Modern American Journal of Engineering, Technology, and Innovation. – 2026. – Vol. 2, No. 3. – P. 18–26.
  8. Fu Y., Wu D., Boulet B. Reinforcement Learning Based Dynamic Model Combination for Time Series Forecasting // Proceedings of the AAAI Conference on Artificial Intelligence. – 2022. – Vol. 36, No. 6. – P. 6639–6647.
  9. Bertsimas D., Boussioux L. Ensemble Modeling for Time Series Forecasting: an Adaptive Robust Optimization Approach // arXiv:2304.04308. – 2023.
  10. Cesa-Bianchi N., Lugosi G. Prediction, Learning, and Games. – Cambridge University Press, 2006.
  11. Kabildjanov A., Pulatov G. Neural Network Forecasting of “Quality of the Day” for the Category of People Sensitive to Weather with Cardiovascular Diseases // Journal of Communications Software and Systems. – 2025. – Vol. 21, No. 4. – P. 496–503.
  12. Silva E.G., Cavalcanti G.D.C., Ren T.I. A Dynamic Predictor Selection Method Based on Recent Temporal Windows for Time
  13. Saadallah A., Jakobs M., Morik K. Online Ensemble Aggregation using Deep Reinforcement Learning for Time Series Forecasting // Proceedings of IEEE DSAA 2021. – 2021. – P. 1–8.
  14. Series Forecasting // IEEE Access. – 2021. – Vol. 9. – P. 108466–108479.