ABSTRACT
The rapid and far-reaching spread of COVID-19 related news can alter investors' perceptions and investment behaviour and affect security returns. We investigate this hypothesis by studying the impact of COVID-19 related news (panic news, media-hype new, fake news, infodemic, and Stringency measures) and alternative measures of COVID-19 developed by Narayan et al. (2021) on Sukuk returns during different market conditions. Empirical results show asymmetric comovement between global Sukuk returns, panic news, travel bans, and the percentage of information about the novel COVID-19 pandemic. Moreover, the COVID-19 news affects the Sukuk returns only when the sukuk markets are bearish. The Sukuk returns are not affected by medical and vaccine information, and the aggregate COVID-19 index that captures the pandemic sentiment.
ABSTRACT
Fake news has been in our society for a long time but with the introduction of social media, Internet and mobile phones the spread of fake news has severely increased. Social media sites are being used to effectively spread misinformation and hoaxes around the world which not only causes people to change their thinking but also manipulates their opinions and decisions. In today’s world it has become nearly impossible to detect if the given news is fake or real. With the arrival of the novel coronavirus-19 pandemic the propagation of fake news is now more than ever. In this time there is a need for something which can classify if a given news is real or not. In this article we aim to develop a model which, using some algorithms, determines if the given news is fake or not. Machine Learning is a form of Artificial Intelligence which utilizes various strategies applying them on data and algorithms to do the same way humans learn. Previous data is used as input by machine learning algorithms to predict new output values. Fake news has the ability to hurt both individuals and society if it is widely spread such as riots, violence and hatred against a community. Understanding the truth of new information and its message can have a positive impact on society when used in conjunction with news detection. We created four prediction models using machine learning that have an accuracy of above 90% which predicts if the given news is either fact or capped (fake). © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.