A New Sentiment Analysis System of Climate Change for Smart City Governance Based on Deep Learning

With the world’s massive population growth, governments over the world are trying to move closer to their citizens, in order to improve the management and governance of their cities, and made it smarter than ever before. Smart city technologies are used to analyze and evaluate huge volumes of data to monitor city dwellers, for better governance. Moreover, social media can be a useful vehicle for governments to better understand their citizens. Twitter sentiment analysis is a great approach to provide deep insight into how citizens behave towards phenomena and thus has definite use for smart city governance and monitoring. Climate change is a critical phenomenon. In recent years, the existence of climate change or global warming became an increasingly public debate. In this paper we developed a deep learning model. Based on convolutional Neural Network (CNN) in order to identify believers and deniers of the climate change phenomenon. And we examined the temporal patterns of climate change discussions on Twitter and its driving factors. Results demonstrate that the developed CNN model successfully identified citizens behavior towards climate change with an overall accuracy of 97% denier and 91% believer. Our model provides improved understanding of factors affecting citizen attitudes on climate change, as an efficient tool for smart city monitoring and governance.