DASTURIY TA'MINOTNI ISHLAB CHIQISH JARAYONIDA SUN'IY INTELLEKT TEXNOLOGIYALARIDAN FOYDALANISH SAMARADORLIGINI OSHIRISH MEXANIZMLARI VA VOSITALARI

Yo‘ldoshev S. S., Muradova F. R. & Muminov B. B.


Annotatsiya

Ushbu tadqiqot ishi zamonaviy dasturiy injiniring sohasida sun'iy intellekt (SI) va mashinaviy o‘qitish (MO) algoritmlarining integratsiyalashuvi jarayonini tahlil qilishga bag‘ishlangan. Maqolada dasturlash jarayonini avtomatlashtirish, kod sifatini oshirish va ishlab chiqish vaqtini qisqartirishda Katta Til Modellari (Large Language Models - LLM) hamda generativ SI vositalarining o‘rni o‘rganilgan. Tadqiqot davomida GitHub Copilot, ChatGPT, Tabnine kabi vositalarning kod yozish tezligi, xatolarni aniqlash aniqligi va dasturchi kognitiv yuklamasiga ta'siri qiyosiy tahlil qilingan. Natijalar shuni ko‘rsatadiki, SI vositalaridan to‘g‘ri foydalanish dasturlash samaradorligini o‘rtacha 30 foizdan 55 foizgacha oshirishi mumkin, ammo bu jarayon "inson nazorati" (human-in-the-loop) tamoyiliga qat'iy rioya qilishni talab etadi. Maqolada, shuningdek, SI tomonidan yaratilgan kodlarning xavfsizlik jihatlari va intellektual mulk masalalari muhokama qilingan.


Adabiyotlar

  1. Chen, M., Tworek, J., Jun, H., Yuan, Q., Pinto, H. P. d. O., Kaplan, J., ... & Zaremba, W. (2021). "Evaluating large language models trained on code". arXiv preprint arXiv:2107.03374.
  2. Xu, F. F., Vasilescu, B., & Neubig, G. (2022). "In-context learning for code generation: An empirical study". Proceedings of the 37th IEEE/ACM International Conference on Automated Software Engineering.
  3. Nguyen, N., & Nadi, S. (2022). "An empirical evaluation of GitHub Copilot's code suggestions". Proceedings of the 19th International Conference on Mining Software Repositories (MSR '22). pp. 1-5. https://doi.org/10.1145/3524842.3528470
  4. Treude, C., & Hata, H. (2023). "AI-Assisted Coding: The demise of the junior developer?". IEEE Software, 40(3), 8-12.
  5. Ziegler, A., Kalliamvakou, E., Li, X. A., Rice, A., Rifkin, D., Simister, S., ... & Aftandilian, E. (2022). "Productivity assessment of neural code completion". Proceedings of the 6th ACM SIGPLAN International Symposium on Machine Programming. pp. 21-29.
  6. OpenAI. (2023). "GPT-4 Technical Report". arXiv preprint arXiv:2303.08774.
  7. Barke, L., James, M. B., & Polikarpova, N. (2023). "Grounded Copilot: How Programmers Interact with Code-Generating Models". Proceedings of the ACM on Programming Languages, 7(OOPSLA1), 1-27.