Research and Development of Fake News Detection Algorithms

Fake news is a rising problem in the modern world. According to a research [24], Americans rate it as a larger problem than racism, climate change, or terrorism. Fake news is becoming a longstanding issue affecting businesses and society, printed and digital media, stock markets and social media. The damage forecasted is expected to be much bigger than any of a kind. Even such a social issue like COVID-19 pandemic is being used for information warfare campaigns. Digital researchers at New York University and Stanford have recently examined this question in a more rigorous way, with some unnerving results. Every day, over a period of several months, they selected five news items that had been published in the past 24 hours and asked different groups of 90 people across the US to ascertain whether or not they were true. They also asked professional fact-checkers to check the same stories to provide some “ground truth” as to their veracity, as Joshua Tucker, a politics professor and co-director of the NYU Center for Social Media and Politics, explained at a recent conference held at the Facebook campus. The preliminary results of this research (which is still under way and was funded by the Hewlett Foundation) show that most participants could tell that true news was true. But, warns Tucker, they were “not good at identifying fake news”. The main goal of the current work is to conduct a research on automated methods of fake news detection and to propose an improved algorithm which than could be used for autonomous fake news detection.