Affiliations School of Industrial Engineering, Purdue University, West Lafayette, Indiana, United States of America, Ecological Sciences and Engineering, Purdue University, West Lafayette, Indiana, United States of America
Auto-generated profile, based on publicly available information
Jackson Bennett
Affiliations School of Industrial Engineering, Purdue University, West Lafayette, Indiana, United States of America, Ecological Sciences and Engineering, Purdue University, West Lafayette, Indiana, United States of America
Surveys are commonly used to quantify public opinions of climate change and to inform sustainability policies. However, conducting large-scale population-based surveys is often a difficult task due to time and resource constraints. This paper outlines a machine learning framework—grounded in statistical learning theory and natural language processing—to augment climate change opinion surveys with social media data. The proposed framework maps social media discourse to climate opinion surveys, allowing for discerning the regionally distinct topics and themes that contribute to climate opinions. The analysis reveals significant regional variation in the emergent social media topics associated with climate opinions. Furthermore, significant correlation is identified between social media discourse and climate attitude. However, the dependencies between topic discussion and climate opinion are not always intuitive and often require augmenting the analysis with a topic’s most frequent n-grams and most representative tweets to effectively interpret the relationship. Finally, the paper concludes with a discussion of how these results can be used in the policy framing process to quickly and effectively understand constituents’ opinions on critical issues.
1 July 2020 in Journal of Management in Engineering
With a mean annual growth rate of roughly 50%, the solar industry has experienced unprecedented growth in the last decade, largely owing to the steadily falling prices of solar installations. Utility-scale energy prices from solar installations are now comparable to all other forms of generation, and the cost of residential system installation has dropped on average by 70%, before incentives. However, the factors explaining this trend extend beyond falling prices. This paper presents a data-centric framework, grounded in machine-learning theory, to estimate solar installations as a function of social, economic, and demographic factors. By doing so, the authors seek to identify the key influencing factors of a community’s adoption of renewable energy. To illustrate the applicability of the proposed data-centric framework, the state of California was selected as a case study. Results indicate that differences in population-adjusted adoption rates can be largely explained by variations in key factors such as income, race, political leaning, average electric power consumption, and solar radiation. By analyzing these differences, decision makers can devise effective incentive mechanisms to nudge homeowners toward improved access to renewable technology.