Abstract
The rapid proliferation of digital platforms and social media has significantly transformed the way information is generated and consumed. However, this transformation has also facilitated the widespread dissemination of fake news, which poses serious threats to societal stability, political systems, and public trust. Fake news refers to deliberately fabricated or misleading information presented as legitimate news, often created to achieve political, financial, or social gains. In recent years, automated fake news detection has emerged as a critical research area leveraging machine learning (ML), deep learning (DL), and multimodal analysis techniques. This paper presents a comprehensive survey of state-of-the-art approaches for fake news detection, including supervised, unsupervised, and semi-supervised learning methods. Furthermore, recent advancements such as transformer-based models, multimodal fusion techniques, and graph-based learning frameworks are discussed.