Fake News Detection: A Comprehensive Review and a Novel Framework

Detecting fake news poses a significant challenge in the area of Natural Language Processing (NLP). The rapid expansion of social networking platforms has not only resulted in a vast amount of information but has also accelerated the spreading of fake news. Consequently, the impact of fake news has grown, sometimes posing a threat to public safety in the real world. This study examines various methodologies for detection of fake news, including Deep Learning (DL), Machine Learning (ML), Transfer Learning, and Hybrid approaches, based on their algorithmic mechanisms. Research papers on fake news detection from the years 2018 to 2024 were collected and analyzed. Initially, these papers were gathered, and an overview of fake news detection, along with its key features, was presented. Subsequently, the available techniques were categorized, and a comprehensive description of each method was provided. The challenges faced by classification methods were also discussed. Based on the analysis a proposed framework has been discussed in the paper.