Social networks have grown quickly in the last few years, and they are now the primary medium used for the publication and spread of false information. This research investigates advanced methodologies for mitigating the proliferation of fake news within the dynamic landscape of social media, employing a focused machine learning paradigm. Leveraging cutting-edge techniques such as natural language processing (NLP), deep learning (DL), and machine learning (ML), our study aims to unravel intricate patterns inherent in deceptive content. In recent times, advancements have been achieved in fake news detection technology, yet numerous challenges persist. While some review papers touch on the subject of fake news detection, this paper aims to provide a more detailed and comprehensive overview, offering a thorough examination of the existing landscape. Rigorous experimentation and evaluation of ML models on diverse datasets by providing a comprehensive understanding of their efficacy in discerning nuanced misinformation strategies. Researchers directed their attention towards analyzing the dissemination patterns of false information within social networks and investigating the correlations among those responsible for spreading such misinformation. The outcomes of this research not only contribute to the arsenal of tools available for cyber security practitioners but also offer insights into the evolving nature of deceptive tactics, aiding in the development of robust, adaptable frameworks for identifying fake news in real-time social media environments. Lastly, the paper reviews different datasets utilized for fake news detection and proposes a new algorithm.