Assessing public perception of climate change: leveraging sentiment analysis to inform effective policy and education strategies in Nigeria

Climate change presents significant challenges to Nigeria, particularly in terms of its impact on agriculture, energy, and water resources. As the country seeks to develop and implement climate adaptation strategies, the application of technology is crucial for building resilience. However, the success of these technological solutions hinges not only on their technical efficacy but also on public acceptance and support. Understanding public sentiment toward climate-smart technologies is essential for ensuring effective adoption and integration into national climate resilience efforts. This study investigates public sentiment toward climate change in Nigeria by analyzing social media discourse, news articles, and online forums through sentiment analysis. The study follows a quantitatively dominant mixed-methods framework combining machine learning sentiment analysis with qualitative thematic interpretation. Data was collected from X (formerly Twitter) and pre-processed using techniques like Time Frequency Inverse Document Frequency (TFIDF) to highlight significant terms, tokenize comments, and calculate inverse document frequency weights. Sentiment analysis was carried out at the document, sentence, and phrase levels using machine learning models and natural language processing (NLP). The study employed two models, Long Short-Term Memory (LSTM) and Support Vector Machine (SVM), which achieved accuracies of 92% and 73%, respectively, in categorizing the sentiments of tweets and comments as positive, negative, or neutral. The results provide insights for government agencies and technology developers to tailor climate adaptation strategies and educational campaigns in alignment with public concerns and expectations. This research contributes to the ongoing effort to build climate resilience in Nigeria by understanding how the public perceives and engages with climate change