Document Type
Dissertation
Disciplines
1.2 COMPUTER AND INFORMATION SCIENCE, Computer Sciences
Abstract
Fake news has become a pervasive issue in today’s society, fuelled by the rapid growth of social media and the changing landscape of news consumption. Detecting and mitigating the spread of fake news is crucial to preserve public trust and protect the integrity of democratic processes. This research highlights the limitations of training models on news datasets that cover various topics and emphasises the importance of using datasets specific to a particular domain for effective detection. The focus of this dissertation is on predicting COVID-19 fake news using style-based methods within a traditional ML framework. Specifically, the study examines the language characteristics of text at different levels. The performance of various Random Forest and XG Boost models are evaluated based on accuracy and F1-score. The results demonstrate the effectiveness of different combinations of features at the lexicon, syntax, semantic, and discourse levels. Lexicon and deep syntax features perform well individually, while discourse features perform poorly. However, combining features leads to strong overall performance. When all language level features are used, the features extracted from the COVID Khan dataset achieve an accuracy of 93.2% using the XG Boost algorithm. Moreover, models trained on COVID-19 specific features outperform those trained on general fake news datasets when predicting COVID-19 fake news, as confirmed by statistical tests. The techniques and findings of this research can be applied to other vulnerable domains to mitigate the harmful impact of fake news.
Recommended Citation
Madden, R. (2023). A Style-Based Approach for Detecting COVID-19 Fake News. [Technological University Dublin].
Creative Commons License
This work is licensed under a Creative Commons Attribution-Share Alike 4.0 International License.
Publication Details
A dissertation submitted in partial fulfilment of the requirements of Dublin Institute of Technology for the degree of M.Sc. in Computing (TU060 - Advanced Software Development) June 14, 2023.