Abstract

This study analyzes Twitter’s contribution to green energy. More than 200,000 global tweets sent during 2020 containing the terms “green energy” OR “greenenergy” were analyzed. The tweets were captured by web scraping and processed using algorithms and techniques for the analysis of massive datasets from social networks. In particular, relationships between users (through mentions) were determined according to the Louvain multilevel algorithm to identify communities and analyze global (density and centralization) and node-level (centrality) metrics. Subsequently, the content of the conversation was subject to semantic analysis (co-occurrence of the most relevant words), hashtag analysis (frequency analysis), and sentiment analysis (using the Vader model). The results reveal nine main communities and their leaders, as well as three main topics of conversation and the emotional state of the digital discussion. The main communities revolve around politics, socioeconomic issues, and environmental activism, while the conversations, which have developed mostly in positive terms, focus on green energy sources and storage, being aligned with the main communities identified, i.e., on political, socioeconomic, and climate change issues. Although most of the conversations have been about socioeconomic issues, the presence of leading company accounts was minor. The main aim of this work is to take the first steps toward an innovative competitive intelligence methodology to study and determine trends within different scientific fields or technologies in society that will enable strategic decisions to be made.