AI FOR URBAN MOBILITY: TRACKING PUBLIC SENTIMENT ON TASHKENT TRANSPORT VIA SOCIAL MEDIA

ТАТУ хабарлари · 2025-yil

Annotatsiya

This paper investigates the use of Natural Language Processing (NLP) to monitor real-time public sentiment about Tashkent's public transportation system through Uzbek (Latin and Cyrillic) and Russian social media posts.Recent policy changes have transformed the city's transport network, but their public reception remains underexplored.Over a two-month period (May-June 2025), we collected 980 relevant public posts from social media.After preprocessing for mixed scripts, informal spellings, and code-switching, we applied sentiment analysis using a fine-tuned multilingual BERT (mBERT) model and topic modeling via Latent Dirichlet Allocation (LDA).Results indicate that negative sentiment dominated (68%), with complaints focused on service delays, overcrowding, malfunctioning payment devices, inaccurate realtime tracking, and frequent air conditioning failures in summer.Positive sentiment (9%) centered on the comfort of new articulated and electric buses and polite driver behavior.Our findings demonstrate the potential of NLP tools to provide continuous, lowcost passenger feedback in low-resource language environments, offering transport authorities actionable insights for targeted service improvements.

Maqola ma’lumotlari
JurnalТАТУ хабарлари
Nashr sanasi2025-09-12
DOI10.61663/252tuitmct2

ТАТУ хабарлари jurnalidan boshqa maqolalar

ТАТУ хабарлари — barcha maqolalar