JBT Journal
Advanced

Raqamli Transformatsiya va Sun’iy Intellekt

TIBBIY TASHXIS QO‘YISH MASALALARINI YECHISHNING NAZARIY ASOSLARI

Eshboyev Erkin Abdirashidovich1, Klicheva Firuza Gulmuratovna1, Daminova Shaxzoda Egamberdi qizi1, Abdirashidov Shohruhmirzo Erkin o’g’li2

1 Qarshi DU

2 O’zMU

Zenodo DOI: 10.5281/zenodo.24365024

Keywords

aralash turdagi alomatlar ko‘p sinfli tasniflash regulyarizatsiya kichik hajmli tanlanma sun’iy intellekt tibbiy tashxis diskriminant tahlil Gauer o‘xshashlik o‘lchovi

Abstract

Maqolada kichik hajmli tibbiy tanlanmalar asosida tashxis qo‘yish masalalarini yechishning nazariy asoslari ko’rib o’tilgan. Ya’ni tibbiy tashxis masalalari sun’iy intellekt nuqtai nazaridan ma’lumot modalligi, bilim manbai, tanlanma hajmi va qaror qabul qilish tartibi bo‘yicha tasniflangan. Tashxis masalasining formal qo‘yilishi xatoliklar qiymati asimmetrik bo‘lgan klassifikator qabul qiladigan qarorlar natijasida kutiladigan o‘rtacha yo‘qotish (bayescha xavf) va Yu.I.Juravlyovning algebraik yondashuvi doirasida berilgan, kichik hajmli tanlanma holati esa obyektlar soni, alomatlar soni va sinflar soni orasidagi munosabatlar orqali aniqlangan. Ko‘p sinfli tashxis masalalarining o‘ziga xosligi, ya’ni sinflar nomutanosibligi, ierarxik tuzilma, “bittaga qarshi qolganlar”, “bittaga qarshi bitta” va dekompozitsiyalar hamda baholash metrikalari tahlil qilingan. Aralash turdagi alomatlar muammosi o‘xshashlik o‘lchovlari nuqtai nazaridan ko‘rib chiqilgan. Ikkita hisoblash tajribasida klassik chiziqli diskriminant tahlili tanlanma hajmi kichraygan sari keskin yomonlashishi (har sinfda 5 ta obyektda makro-F1 = 0,744), Ledoit–Wolf qisqartirishi esa uni 0,934 gacha tiklashi, aralash turdagi ma’lumotlarda alomatlar turini hisobga olmaydigan Evklid masofasi makro-F1 qiymatini 0,362 gacha tushirib yuborishi ko‘rsatib o’tilgan. Olingan natijalar kichik hajmli tibbiy ma’lumotlar uchun regulyarizatsiyalangan va alomatlar turini hisobga oluvchi usullarni tanlash bo‘yicha nazariy asos bo‘lib xizmat qiladi.

Views: 33 Published: 2026-10-28 Pages: 166-175 Volume: 4 Issue: 5-Special-Issues

References

  1. Aeberhard S., Coomans D., de Vel O. Comparative analysis of statistical pattern recognition methods in high dimensional settings // Pattern Recognition. - 1994. - Vol. 27, No. 8. - P. 1065–1077.
  2. Bishop C. M. Pattern Recognition and Machine Learning. - New York: Springer, 2006. - 738 p.
  3. Raudys S. J., Jain A. K. Small sample size effects in statistical pattern recognition: recommendations for practitioners // IEEE Transactions on Pattern Analysis and Machine Intelligence. - 1991. - Vol. 13, No. 3. - P. 252–264.
  4. Gower J. C. A general coefficient of similarity and some of its properties //
  5. Журавлёв Ю. И. Об алгебраическом подходе к решению задач распознавания или классификации // Проблемы кибернетики. - 1978. - Вып. 33. - С. 5–68.
  6. Chawla N. V., Bowyer K. W., Hall L. O., Kegelmeyer W. P. SMOTE: synthetic minority over-sampling technique // Journal of Artificial Intelligence Research. - 2002. - Vol. 16. - P. 321–357.
  7. Ledoit O., Wolf M. A well-conditioned estimator for large-dimensional covariance matrices // Journal of Multivariate Analysis. - 2004. - Vol. 88, No. 2. - P. 365–411.
  8. Hand D. J., Till R. J. A simple generalisation of the area under the ROC curve for multiple class classification problems // Machine Learning. - 2001. - Vol. 45, No. 2. - P. 171–186.
  9. Little R. J. A., Rubin D. B. Statistical Analysis with Missing Data. - 3rd ed. - Hoboken: Wiley, 2019. - 462 p.
  10. Rifkin R., Klautau A. In defense of one-vs-all classification // Journal of Machine Learning Research. - 2004. - Vol. 5. - P. 101–141.
  11. Dietterich T. G., Bakiri G. Solving multiclass learning problems via error-correcting output codes // Journal of Artificial Intelligence Research. - 1995. - Vol. 2. - P. 263–286.
  12. Tibshirani R. Regression shrinkage and selection via the lasso // Journal of the Royal Statistical Society: Series B. - 1996. - Vol. 58, No. 1. - P. 267–288.
  13. Vabalas A., Gowen E., Poliakoff E., Casson A. J. Machine learning algorithm validation with a limited sample size // PLoS ONE. - 2019. - Vol. 14, No. 11. - e0224365.
  14. Varma S., Simon R. Bias in error estimation when using cross-validation for model selection // BMC Bioinformatics. - 2006. - Vol. 7. - Art. 91.
  15. Allwein E. L., Schapire R. E., Singer Y. Reducing multiclass to binary: a unifying approach for margin classifiers // Journal of Machine Learning Research. - 2000. - Vol. 1. - P. 113–141.
  16. Esteva A., Kuprel B., Novoa R. A. et al. Dermatologist-level classification of skin cancer with deep neural networks // Nature. - 2017. - Vol. 542. - P. 115–118.
  17. Журавлёв Ю. И., Рязанов В. В., Сенько О. В. «Распознавание». Математические методы. Программная система. Практические применения. - М.: Фазис, 2006. - 176 с.
  18. Hastie T., Tibshirani R., Friedman J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. - 2nd ed. - New York: Springer, 2009. - 745 p.
  19. Rajkomar A., Dean J., Kohane I. Machine learning in medicine // New England Journal of Medicine. - 2019. - Vol. 380, No. 14. - P. 1347–1358.
  20. Grinsztajn L., Oyallon E., Varoquaux G. Why do tree-based models still outperform deep learning on typical tabular data? // Advances in Neural Information Processing Systems 35 (NeurIPS 2022), Datasets and Benchmarks Track. - 2022.
  21. Vapnik V. N. Statistical Learning Theory. - New York: Wiley, 1998. - 736 p.
  22. Shortliffe E. H. Computer-Based Medical Consultations: MYCIN. - New York: Elsevier, 1976. - 264 p.
  23. Biometrics. - 1971. - Vol. 27, No. 4. - P. 857–871.
  24. Topol E. J. High-performance medicine: the convergence of human and artificial intelligence // Nature Medicine. - 2019. - Vol. 25. - P. 44–56.
  25. Ledley R. S., Lusted L. B. Reasoning foundations of medical diagnosis // Science. - 1959. - Vol. 130, No. 3366. - P. 9–21.