O‘ZBEK TILIDAGI MATNLARNI SENTIMENT TAHLIL QILISH VA AVTOMATIK TASNIFLASH USULLARINI ISHLAB CHIQISH
Keywords:
Kalit so‘zlar: Sentiment tahlil, opinion mining, tabiiy tilni qayta ishlash, NLP, ijobiy matn, salbiy matn, neytral matn, lingvistik tahlil, morfologiya, o‘zbek tili, matn tasnifi., Keywords: Sentiment analysis, opinion mining, natural language processing, NLP, positive text, negative text, neutral text, linguistic analysis, morphology, Uzbek language, text classification.Abstract
Annotatsiya: Ushbu tezisda o‘zbek tilidagi matnlarni sentiment tahlil qilish masalasi ko‘rib chiqiladi. Sentiment tahlil matnlardagi hissiy munosabatni aniqlab, ularni ijobiy, salbiy va neytral toifalarga ajratadi. Tadqiqotda turli matnlar lingvistik va statistik yondashuvlar asosida tahlil qilindi. Natijalarga ko‘ra, neytral matnlar ustunlik qilishi, ijobiy va salbiylik esa asosan leksik birliklar orqali ifodalanishi aniqlandi. Shuningdek, o‘zbek tilining morfologik murakkabligi avtomatik tahlil jarayoniga ta’sir ko‘rsatishi kuzatildi.
Abstract: This thesis examines the problem of sentiment analysis of Uzbek-language texts. Sentiment analysis aims to identify the emotional polarity in texts and classify them into positive, negative, and neutral categories. In this study, various texts were analyzed using both linguistic and statistical approaches. The results indicate that neutral texts are predominant, while positive and negative sentiments are mainly expressed through lexical units. It was also observed that the morphological complexity of the Uzbek language significantly affects the automatic text analysis process.
