Information and communication and chemical technologies

Vol. 1 No. 30 (2026) / 2026-03-30 / Number of views: 223

COMPUTATIONAL MODELING OF PUBLIC PERCEPTION OF ARTIFICIAL INTELLIGENCE ON REDDIT USING ML

Authors

Kazakh-British Technical University
https://orcid.org/0009-0005-6465-1325
Ravensbourne University London
https://orcid.org/0000-0002-3117-3997
Sh.Yessenov Caspian University of Technology and Engineering
https://orcid.org/0000-0003-0592-5865

Keywords

artificial intelligence, generative AI, sentiment analysis, topic modeling, social media mining, public perception, natural language processing

Link to DOI:

https://doi.org/10.58805/kazutb.v.1.30-1540

How to quote

Algashbekova, N., et al. “COMPUTATIONAL MODELING OF PUBLIC PERCEPTION OF ARTIFICIAL INTELLIGENCE ON REDDIT USING ML”. Vestnik KazUTB, vol. 1, no. 30, Mar. 2026, doi:10.58805/kazutb.v.1.30-1540.

Abstract

Discussions about artificial intelligence (AI), particularly generative models like ChatGPT, have become more common on social media these days, reflecting a wide range of public opinions and worries. From 2022 to 2025, this study examines public opinion and thematic discussion on Reddit related to artificial intelligence (AI). Using more than 35 carefully chosen keywords, we gathered more than 2900 posts and 636000 comments from 10 subreddits pertaining to artificial intelligence. Logistic Regression, SVM, and Naive Bayes were used to perform a supervised sentiment classification using a labeled Reddit dataset. Applying the best-performing model to our entiredataset showed that 20% of comments were negative, 35% were neutral, and 45% of comments were positive. BERTopic topic modeling revealed dominant themes like creative tools, automation anxiety, and AI ethics. Our study provides insight into how online communities perceive and respond emotionally to the rise of generative AI.