Online Guard: Identifying the misinformation in social media and its impact on COVID-19 vaccination progress in different countries

Abstract

The emergence of the novel coronavirus pandemic has caused a myriad of problems worldwide. One such problem is misinformation, which in itself should be considered a risk. Since the outbreak of the COVID-19 pandemic, popular social media platforms are flooded by exaggerated phony news which is affecting our society, well-being and public safety. Many of the online falsehoods don't have apparent sources or intentions, rather, some niche groups often start mobilizing to endorse their agendas through the rumors. Although the pertinent tools and existing techniques can support fact-checking and identification of conspiracy, misinformation and negative sentiment at various stages, a complete end-to-end solution is complicated. In this paper, we propose a thorough analysis and identification system named Online Guard using natural language processing tools and supervised learning techniques to identify the relationship between misinformation from the negative sentiment of COVID-19 vaccine-related tweets and vaccination progress rate and its impact in different countries. For this purpose, we will use a COVID-19 all vaccines tweet dataset to identify and analyze misinformation, and another dataset named country vaccination that shows vaccine rollout and vaccination progress in different countries. The aim of this project is to identify the relationship between spreading misinformation, negative emotions on Twitter, and its impact on vaccination progress for a particular time period.