False information about the COVID-19 pandemic’s behaviour and effects, the effects of lockdowns, quarantines, and social isolation, as well as the medical response in terms of vaccines and medications, has been pervasive. In fact, some international leaders misled the public with absurd claims and advice on how to deal with the SARS-CoV-2 issue.
A computer chatbot that uses the ensemble learning technique to identify fake news is discussed in research published in the International Journal of Artificial Intelligence and Soft Computing, which is relevant given that the epidemic is still very much a problem. A team from the University of Delhi in India created the chatbot, which they named CovFakeBot. It was taught using well-known machine-learning algorithms. When it comes to the COVID-19 pandemic, it can accurately distinguish between real news and false information on the microblogging site Twitter.
The Shaheed Rajguru College of Applied Sciences for Women’s Department of Computer Science’s Hunar Batra, Gunjan Kanwar Palawat, Kanika Gupta, Priadarshana, Supragya, Deepali Bajaj, and Urmil Bharti explain that their chatbot combines Twilio with the application programming interface (API) of a well-known messaging app, specifically the WhatsApp Business API, to produce a conversational user interface. On a dataset from Twitter, the team tested the chatbot using ten different machine learning and ensemble learning classifiers. The most realistic model was found to be a soft-voting one.
The CovFakeBot, according to the team, could end up being a very helpful tool for social media users who want to swiftly determine whether a worrying item is real news or fake news. They note that by training a new instance of the chatbot with fresh data relevant to the area of interest, it would be reasonably easy to expand the system to other areas where fake news is a problem. They anticipate that CovFakeBot and its relatives will help to curtail the propagation of false information on social media in the long run.
