The rapid spread of fake news on social media platforms poses serious threats to public trust, democratic processes, and social stability. Manual verification approaches are insufficient due to the scale and velocity of online content, necessitating automated solutions. This paper presents a comprehensive review of machine learning (ML) and deep learning (DL) approaches for fake news detection and proposes an ensemble framework combining Support Vector Machine (SVM) and Deep Neural Network (DNN) models. Thirty-four peer-reviewed studies are ana-lyzed to identify trends, performance benchmarks, and research gaps. Experimental evaluation on multiple benchmark datasets demonstrates that the proposed ensemble achieves an accuracy of 94.6% and macro-F1 score of 0.946, outperforming individual classifiers. The findings highlight the importance of robust preprocessing, dataset diversity, and hybrid learning models for reliable misinformation detection. Future work emphasizes multimodal learning, transformer architec-tures, and real-time deployment in large-scale social media systems.
Bajgai et al. (Thu,) studied this question.