Development of a COVID-19 Misinformation Detection System using Machine Learning and Natural Language Processing Techniques

Óscar Parro Sainz. (2023). Development of a COVID-19 Misinformation Detection System using Machine Learning and Natural Language Processing Techniques. Final Career Project (TFG). Universidad Politécnica de Madrid, ETSI Telecomunicación.

Abstract:
Since the appearance of COVID in 2019, not only this virus has spread rapidly, but also a lot of misinformation has been transmitted worldwide by news or internet users. This fake news has caused not only a lack of information, but also disruption in society and even deadly consequences in health problems. Therefore, being able to understand and mitigate such misinformation about COVID-19 is not only of academic interest, but also of enormous social impact. In the early years of the COVID-19 spread, most of the society did not really understand the virus and how to act against it. However, over the years, information and data has increased, broadening experts’ knowledge about the virus. Even so, many media or Internet users continue to disseminate false information. Behind these actions, strategies are articulated on numerous occasions to manipulate public opinion and erode the stability of States and their institutions. The main objective of this project, within the field of Artificial Intelligence, is to use Natural Language Processing (NLP) and Machine Learning (ML) techniques to generate a system for the automatic detection of fake news in relation to COVID. The system developed will be used to scan all the information found on the different media sources, such as news, articles or social media posts related to COVID-19. Being able to analyse misinformation trends found in these and determine what information is most likely to be false. The implementation of a fake news detection system could play a crucial role in the fight against misinformation in the context of COVID-19. By providing an efficient and accurate way of identifying and alerting about fake news, it can help minimise its negative impact on society. In addition to its usefulness in detecting fake news, the developed system could have a significant impact on promoting media literacy and education in the responsible use of information. By analysing and categorising the veracity of news or social media posts, users could be made more aware of the tactics and strategies used to spread false information. This would enable people to make more informed and critical decisions when consuming COVID-19 related content.