COVID-19 fake news detection model on social media data using machine learning techniques

• Main Authors: Kai Xuan, Kelvin Liew, Bhuiyan, Mohaiminul Islam, Nur Shazwani, Kamarudin, Ahmad Fakhri, Ab Nasir, Muhammad Zulfahmi, Toh Abdullah@ Toh Chin Lai
• Format: Conference or Workshop Item
• Language: English; English
• Published: Institute of Electrical and Electronics Engineers Inc. 2023
• Subjects: QA75 Electronic computers. Computer science; QA76 Computer software; T Technology (General); TA Engineering (General). Civil engineering (General)
• Online Access: http://umpir.ump.edu.my/id/eprint/40374/1/COVID-19%20fake%20news%20detection%20model%20on%20social%20media.pdf; http://umpir.ump.edu.my/id/eprint/40374/2/COVID-19%20fake%20news%20detection%20model%20on%20social%20media%20data%20using%20machine%20learning%20techniques_ABS.pdf; http://umpir.ump.edu.my/id/eprint/40374/; https://doi.org/10.1109/ICSECS58457.2023.10256386

Social media sites like Instagram, Twitter, and Facebook have become indispensable parts of the daily routine. These social media sites are powerful instruments for spreading the news, photographs, and other sorts of information. However, since the emergence of the COVID-19 pandemic in December 2019, many articles and headlines concerning the COVID-19 epidemic have surfaced on social media. Social media is frequently used to disseminate fraudulent material or information. This disinformation may confuse consumers, perhaps causing worry. It is hard to counter the widespread dissemination of disinformation. As a result, it is critical to develop a model for recognizing fake news in the news stream. The dataset, which would be a synthesis of COVID-19-related news from numerous social media and news sources, is utilized for categorization in this work. Markers are retrieved from unstructured textual data gathered from a variety of sources. Then, to eliminate the computational burden of analyzing all of the features in the dataset, feature selection is done. Finally, to categorize the COVID -19 related dataset, multiple cutting-edge machine-learning algorithms were trained. Support Vector Machine (SVM), Naïve Bayes (NB), and Decision Tree (DT) are the machine learning models presented. Finally, numerous measures are used to evaluate these algorithms such as accuracy, precision, recall, and F1 score. The Decision Tress algorithm reported the highest accuracy of 100% compared to the Support Vector Machine 98.7% and Naïve Bayes 96.3%.