ANALYZING MEDIA NARRATIVES ON RENEWABLE ENERGY IN MACEDONIA USING SENTIMENT AND TOPIC MODELING

Fri, 26 Dec 2025 00:00:00 +0100

This paper explores the evolution of media sentiment and thematic discourse around Renewable Energy Sources (RES) in Macedonia from 2017 to 2024. Drawing on over 9,500 news teasers retrieved from Time.mk, we apply FinBERT and BERTweet for sentiment analysis and Latent Dirichlet Allocation (LDA) for topic modeling. Our findings reveal a dominant neutral sentiment profile with notable positive sentiment peaks aligned with key regulatory developments and market shifts. Topic modeling further captures the transformation of discourse from centralized, policy-driven rhetoric toward decentralized, technologically mature themes, centered on solar expansion, public engagement, and institutional implementation. The outcome of this research contributes to understanding how public discourse on RES evolves over time, and its implications add value to the role of media in shaping societal perception and stakeholder’s response to renewable energy transition.

CONTEMPORARY MATERIALS

ANALYZING MEDIA NARRATIVES ON RENEWABLE ENERGY IN MACEDONIA USING SENTIMENT AND TOPIC MODELING