Vol. 3 No. 6 (2025), Articles
Vol. 3 No. 6 (2025)

PROGMALINGVISTIKA: SUN’IY INTELLEKT TILLARNI QANDAY O‘RGANMOQDA VA INSON TILINI QANDAY O‘ZGARTIRMOQDA

Articles
Published 2025-11-26
Duschanova Shoxida Baxramovna
PDF

Keywords

progmalingvistika, prompt engineering, sun’iy intellekt, katta tilli modellar (LLM), Chain-of-Thought, o‘zbek tili AI, agglutinativ tillar, tilshunoslik va AI, avtomatik prompt optimallashtirish, Claude, GPT-4o, Grok.

Abstract

Progmalingvistika — sun’iy intellekt tilli modellari (LLM) bilan inson tili orqali eng samarali muloqot qilish san’ati va ilmiy yo‘nalishidir. Ushbu maqola progmalingvistikaning 2022–2025 yillardagi jadal rivojlanishini, iqtisodiy ahamiyatini, eng samarali prompt texnikalarini va tilshunoslik bilan chambarchas bog‘liqligini tahlil qiladi. Maqola 120 dan ortiq ilmiy ish, OpenAI, Anthropic, Google va xAI ochiq hisobotlari, 1 200 nafar professional prompt-muhandis ishtirokidagi so‘rovnoma hamda o‘zbek tilidagi 15 ta mahalliy model ustida o‘tkazilgan empirik tajribalarga asoslanadi. Natijalar shuni ko‘rsatadiki, to‘g‘ri progmalingvistik yondashuv javob sifatini 70–117 % gacha oshirishi, o‘zbek kabi agglutinativ tillarda esa maxsus morfologik eslatmalar xatolik darajasini 10 baravar kamaytirishi mumkin. Progmalingvistika endi tilshunoslikning yangi bo‘limi sifatida pragmatika, semantika va sotsiolingvistika masalalarini qayta shakllantirmoqda.

PDF

References

1. Wei, J. et al. (2022). Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. NeurIPS 2022.

2. Yao, S. et al. (2023). Tree of Thoughts: Deliberate Problem Solving with Large Language Models. NeurIPS 2023.

3. Kojima, T. et al. (2024). Large Language Models are Zero-Shot Reasoners (Updated 2024). arXiv:2403.14487.

4. Wang, X. et al. (2022). Self-Consistency Improves Chain of Thought Reasoning in Language Models. ICLR 2023.

5. Trotsenko, D. et al. (2025). Morphology-Aware Prompting for Agglutinative Languages: Uzbek, Turkish, Finnish. ACL 2025 (accepted).

6. OpenAI (2024). System Prompt Leak Collection & Analysis. GitHub: openai-system-prompts-2024.

7. Anthropic (2024). Claude 3.5 System Prompt (14 287 tokens). Public leak, September 2024.

8. Latent Space & Scale AI (2024). Prompt Engineering Salary & Skills Survey 2024 (n=1 200).

9. McKinsey Global Institute (2023). The Economic Potential of Generative AI: The Next Productivity Frontier.

10. Klarna AI Assistant Case Study (2024). “How we replaced 700 customer service agents with one AI”. Klarna Investor Report, Feb 2024.