Characterizing the 2022- Russo-Ukrainian Conflict Through the Lenses of Aspect-Based Sentiment Analysis: Dataset, Methodology, and Key Findings

Abstract

Online social networks (OSNs) play a crucial role in modern society by supporting free expression, information sharing, and social movement organization. However, they are also the tool of choice to spread disinformation, hate speech, and support propaganda. As such, it is crucial to analyze OSNs, particularly during critical events such as elections, pandemics, and conflicts, when disinformation campaigns may seek to undermine the democratic values of a nation. This paper analyzes the general-public perception of the first phases of the 2022- Russo-Ukrainian conflict on Twitter. To this end, we developed a general methodology consisting of several steps. We built a dataset of 5.5+ million tweets related to the subject, generated by 1.8+ million unique users. Then, we cluster users into five categories, and combining statistical analysis and aspect-based sentiment analysis (ABSA), we quantitatively and qualitatively investigate the spread of information during the conflict. Our analysis revealed several important insights, including anomalies in the behavior of specific user categories and their sentiment trends and a spike in the daily account creation rate before the conflict. Other than being interesting on their own, our findings also have significant implications for future research on how disinformation campaigns are executed and on developing effective strategies to mitigate their impact.