SUN’IY INTELLEKT MODELLARI YORDAMIDA KASBIY MOYILLIKNI BASHORAT QILISH VA ULARNING QIYOSIY TAHLILI
Keywords:
Kalit so'zlar: Sun'iy intellekt, Machine Learning, Katta til modellari (LLM), RIASEC, kasbiy yo'naltirish, XGBoost.Abstract
Annotatsiya: Ushbu tezis shaxsning psixometrik ko'rsatkichlari (xususan,
Hollandning RIASEC modeli) asosida unga eng mos kasbiy faoliyat turini aniqlashda
sun'iy intellekt modellarini qo'llash masalalariga bag'ishlangan. Bugungi jadal
o'zgaruvchan mehnat bozorida yoshlarni ularning layoqati va qiziqishlariga mos
ravishda to'g'ri kasb-hunarga yo'naltirish o'ta dolzarb ahamiyat kasb etmoqda.
Tadqiqotda klassik mashinali o'rganish algoritmlari (XGBoost, Random Forest,
Logistic regression, KNN va Decision tree) hamda ilg'or katta til modellari (LLM –
Gemini) qiyosiy tahlil qilinib, ularning ishonchliligi 63 000 dan ortiq real so'rovnoma
bazasi va O'zbekiston sharoitida to'plangan ma'lumotlar asosida isbotlangan.
References
Foydalanilgan adabiyotlar:
1. Holland, J. L. (1997). Making vocational choices: A theory of vocational
personalities and work environments. Psychological Assessment Resources.
2. Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. KDD
'16.
3. Open Psychometrics Project Data. [Manba: openpsychometrics.org]
4. Naveed, H., et al. (2023). A Comprehensive Overview of Large Language Models.
