Detecting Fake News Spreaders on Twitter Through Follower Networks

Obtaining news from social media platforms has become increasingly popular due to their ease of access and high speed of information dissemination. These same factors have, however, also increased the range and speed at which misinformation and fake news spread. While machine-run accounts (bots) contribute significantly to the spread of misinformation, human users on these platforms also play a key role in contributing to the spread. Thus, there is a need for an in-depth understanding of the relationship between users and the spread of fake news. This paper proposes a new data-driven metric called User Impact Factor (UIF) aims to show the importance of user content analysis and neighbourhood influence to profile a fake news spreader on Twitter. Tweets and retweets of each user are collected and classified as ‘fake’ or ‘not fake’ using Natural Language Processing (NLP). These labeled posts are combined with data on the number of the user’s followers and retweet potential in order to generate the user’s impact factor. Experiments are performed using data collected from Twitter and the results show the effectiveness of the proposed approach in identifying fake news spreaders.