Date of Award
Master of Science
Geography and Environmental Resources
Energy has been at the top of the national and global political agenda along with other concomitant challenges, such as poverty, disaster and climate change. Social perception on various energy issues, such as its availability, development and consumption deeply affect our energy future. This type of information is traditionally collected through structured energy surveys. However, these surveys are often subject to formidable costs and intensive labor, as well as a lack of temporal dimensions. Social media can provide a more cost-effective solution to collect massive amount of data on public opinions in a timely manner that may complement the survey. The purpose of this study is to use machine learning algorithms and social media conversations to characterize the spatiotemporal topics and social perception on different energy in terms of spatial and temporal dimensions. Text analysis algorithms, such as sentiment analysis and topic analysis, were employed to offer insights into the public attitudes and those prominent issues related to energy. The results show that the energy related public perceptions exhibited spatiotemporal dynamics. The study is expected to help inform decision making, formulate national energy policies, and update entrepreneurial energy development decisions.
This thesis is Open Access and may be downloaded by anyone.