The Sentiment of Fake News

Over the past few years, extensive research has been conducted on the topic of Fake News across various social networks, exploring several aspects and disciplines. There are two main issues concerning Fake News in social networks. The identification of Fake and Real news, and the tracking and prevention of their dissemination on social media, while identifying users, organizations, or pages as Fake News propagators. Among the prominent platforms facilitating the propagation of Fake News, the platform once known as Twitter stands out as a widespread medium. Different users, public figures and organizations use it to shape public opinion and garner support, often employing problematic and inaccurate data. This study focuses on a specific aspect of Fake News, namely their sentiment. Previous research and prevailing public opinion suggest that Fake News tends to exhibit a predominantly negative sentiment. Based on this initial assumption, we use NLP techniques of sentiment analysis, to build a sentiment analyzer and use it on large scale datasets of verified mixed Fake and True news. The findings provide interesting insights into the different sentiment of Fake News when the topics under consideration are different.