Enhancing the Detection of Fake News in Social Media: A Comparison of Support Vector Machine Algorithms, Hugging Face Transformers, and Passive Aggressive Classifier

In this research, we compare and contrast many different AI algorithms designed to improve the ability to spot false information on social media. In particular, it assesses the efficacy of Passive Aggressive Classifier (Crammer et al., J Mach Learn Res 7:551–585, 2006), Hugging Face (Devlin et al., Bert: pre-training of deep bidirectional transformers for language understanding. Cornell University, 2018), and Support Vector Machine Algorithms (Hearst et al., IEEE Intel Syst Appl 13:18–28, 1998; Boser et al., Proceedings of fifth annual workshop computational learning theory, pp 144–152, 1992; Schlkopf et al., Advances in kernel methods support vector learning. Mass, Cambridge, 1998; Cristianini and Shawe-Taylor, An introduction to support vector machines and other kernel-based learning methods. Cambridge University Press, 2000; Baarir et al., Algeria 2021:125–130, 2020). Amid the increasing menace of misinformation on social media, the need for effective fake news detection mechanisms cannot be overstated. The study begins with an overview of the algorithms under review, followed by an explanation of their application in fake news detection. This analysis then moves into a comparison mode, assessing each method according to several criteria including computational complexity, accuracy, precision, and recall. The research goes further into the pros and cons of each model, illuminating how well they perform with various sets of data and varieties of disinformation. In order to create reliable and accurate false news detection systems, it is important to determine which algorithms are the most successful. The results of this comparison not only add to the body of knowledge on disinformation identification, but they also provide concrete strategies for bolstering the trustworthiness of content shared on social media.