ENSURING THE RELIABILITY OF SCIENTIFIC RESULTS BASED ON EXPLAINABLE AI (XAI)
Keywords:
Explainable AI, XAI, LIMEAbstract
This paper investigates the role of Explainable AI (XAI) technologies in analyzing scientific results and enhancing their reliability in natural and exact sciences. The study employs modern XAI methods such as LIME and SHAP to assess model transparency, error rates, and alignment with expert evaluations. Results indicate that XAI methods improve model reliability by 23–30%, reduce incorrect decisions, and strengthen trust in scientific research. The findings demonstrate the practical applicability of XAI approaches as an effective tool in scientific studies. The development of automated scientific platforms based on XAI is considered a relevant task for future research and practice.References
1. Wickramasinghe, C. S., Marino, D., & Amarasinghe, K. (2023). Explainable artificial intelligence. Frontiers in Computer Science, 5:1291752 — XAI sohasidagi umumiy sharh va muhim kontseptlar.
2. Saarela, M., & Podgorelec, V. (2024). Recent Applications of Explainable AI (XAI): A Systematic Literature Review. Applied Sciences, 14(19):8884 — XAI ning so‘nggi qo‘llanilishi bo‘yicha tizimli tahlil.
3. Schneider, J. (2024). Explainable Generative AI (GenXAI): a survey, conceptualization, and research agenda. Artificial Intelligence Review — Generativ AI va Explainability bo‘yicha yuzaga kelayotgan yondashuvlar.
4. Explainable Artificial Intelligence: Second World Conference, xAI 2024, Valletta, Malta, July 17–19, 2024, Proceedings (2024). Springer — XAI bo‘yicha xalqaro konferensiya materiallari.
5. Explainable Artificial Intelligence: Second World Conference, xAI 2024, Valletta, Malta, July 17–19, 2024, Proceedings, Part III. Springer (2024) — XAI konferensiyasi maqolalari,
metodologiyalar va ilovalar.
6. Altaqhi, Z. M., Pradhan, S., & Aljohani, N. (2025). A Systematic Literature Review of the Latest Advancements in XAI. Technologies, 13(3):93 — XAI sohasidagi 2014–2024 yillardagi yutuqlar va istiqbollar.
7. Ghasemi, A., Hashtarkhani, S., Schwartz, D. L., & Shaban-Nejad, A. (2024). Explainable artificial intelligence in breast cancer detection and risk prediction. arXiv — XAI ni tibbiyot sohasida qo‘llash bo‘yicha tizimli skopinig sharhi.
8. Baniecki, H., & Biecek, P. (2023). Adversarial attacks and defenses in explainable artificial intelligence: A survey. arXiv — XAI metodlaridagi tahdidlar va mudofaa strategiyalari.
9. Nauta, M., Trienes, J., Pathak, S., Nguyen, E., et al. (2023). From Anecdotal Evidence to
Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI. ACM
Computing Surveys — XAI baholash mezonlari va metodlar bo‘yicha tizimli sharh.
10.Explainable and interpretable machine learning and data mining. Data Mining and
Knowledge Discovery, 38 (2024) — XAI va interpretatsiya qilingan mashinalarni o‘rganish bo‘yicha
mehmon muharrirlik maqolasi.
