Public perception on climate change: Analyzing impact of IPCC reports via topic modeling with NMF and sentiment analysis

Social engagement in environmental issues has significantly increased with the rise of social media, which plays a critical role in shaping public perception of climate change. This study examines the impact of the Intergovernmental Panel on Climate Change (IPCC) climate assessment reports on Twitter. A total of 897,562 English-language tweets related to the AR4, AR5, and AR6 reports were collected from 2006 to 2022. After preprocessing using Natural language processing (NLP) techniques, 583,493 tweets were analyzed through topic modeling based on Non-negative matrix factorization (NMF) and its nonlinear extension, linear-quadratic NMF (LQ-NMF). The results revealed notable shifts in public discourse before and after the reports' release, highlighting recurring topics such as climate risk, urgency for action, and greenhouse gas emissions. Additionally, sentiment analysis using the VADER method showed a decrease in neutral tweets and greater polarization in responses, especially regarding the AR6 report. These findings suggest that IPCC publications have generated measurable influence on social media and that the applied methods are effective tools for assessing public perception of climate communication.