SENTIMENT TAHLILI ALGORITMLARINING UMUMIY TASNIFI VA ULARNI TAHLIL QILISH
Keywords:
sentiment tahlili, Lexicon-based, Machine Learning, Deep Learning, Transformer, Large Language Models, Hybrid model, matn klassifikatsiyasi, NLP, sun’iy intellekt.Abstract
Mazkur maqolada sentiment tahlilida qo‘llaniladigan algoritmlarning umumiy tasnifi hamda ularning nazariy va amaliy xususiyatlarini tahlil qilinganman. Tadqiqotda sentiment tahlili algoritmlari oltita asosiy guruhga ajratilgan: lug‘atga asoslangan (Lexicon-based), mashinali o‘qitish (Machine Learning), chuqur o‘rganish (Deep Learning), Transformer modellar, katta til modellari (Large Language Models – LLM) hamda gibrid (Hybrid) yondashuvlar. Tadqiqot natijalari turli algoritmlar o‘rtasidagi farqlarni aniqlash, ularning qo‘llanish chegaralarini baholash hamda muayyan amaliy vazifaga mos sentiment tahlili modelini tanlash uchun ilmiy-metodik asos yaratadi. Xususan, gibrid yondashuvlarning klassik usullar va zamonaviy Transformer yoki LLM modellarining afzalliklarini birlashtirishi natijasida aniqlik, barqarorlik va umumlashtirish qobiliyatini sezilarli darajada oshirish imkoniyati mavjudligi asoslab berilgan..
References
1. Kotelnikova A., Paschenko D., Bochenina K., Kotelnikov E. Lexicon-based Methods vs. BERT for Text Sentiment Analysis. arXiv preprint, arXiv:2111.10097, 2021.
2. De S., Gosai A., Thankachan K., et al. From Lexicons to Large Language Models: A Comprehensive Survey of Sentiment Analysis Methods, Benchmarks, and Emerging Frontiers. Computer Modeling in Engineering & Sciences, Vol. 147, No. 2, 2026.
3. Minaee, S., Kalchbrenner, N., Cambria, E., Nikzad, N., Chenaghlu, M., & Gao, J. (2021). Deep Learning–Based Text Classification: A Comprehensive Review. ACM Computing Surveys, 54(3), 1–40.
4. Minaee, S., Cambria, E., Gao, J., et al. (2024). Large Language Models and Transformers for Natural Language Processing: A Survey. ACM Computing Surveys.
5. Zhao, W. X., Zhou, K., Li, J., Tang, T., Wang, X., Hou, Y., et al. (2023). A Survey of Large Language Models. arXiv preprint, arXiv:2303.18223.


