Abstract:
Social media has played the amplifier role of spreading misinformation that it turned into an Infodemic. Researchers utilized natural language processing and machine learning to detect and label misinformation on the network. Literature on fake news detection for content in English and Latin languages using natural language is quite established however the research on Arabic Fake news detection is scarce. This paper contributes to the Arabic fake news detection research by comparing the effectiveness of several machine learning in classifying misinformed tweets based on linguistic features. For this experiment, A hybrid dataset with balanced labels was compiled from sources in the literature. Results reinforce the potential of applications of machine learning in aiding the fact-checking process. Further research in building language resources and classification models is needed for the advancement of Arabic FND applications and their associated fields.
