Pattern Recognition and Image Analysis Computer Vision and Pattern Recognition
Environmental Discourse in Russian Online Communities: Insights from Topic Modeling and Topic Labeling
Abstract
Abstract This paper investigates the content of Russian online communities dedicated to environmental issues. Social media platforms serve as key channels for information dissemination and public mobilization, providing a unique source for analyzing how environmental topics are represented in public debates. The study aims to trace the main directions of environmental discourse and to understand how digital communication reflects the spread of ecological knowledge and practices. To achieve this, state-of-the-art natural language processing (NLP) methods are applied to a large corpus of posts from environmental internet communities. Furthermore, the study compares how human experts and large language models (LLMs) assign labels to identified topics, allowing for an assessment of the potential of computational tools to support research in this field. This work contributes to both the social study of environmental practices in Russia and to the development of text analysis techniques for large online collections. The findings demonstrate the significance of combining expert knowledge with automated approaches to study complex social and ecological processes.