Identifying False News Using Enhanced Passive Aggressive Classifier Methods

There were fake news sources and hoaxes even before the Internet. Most people agree that stories that are made up on the Internet with the goal of deceiving readers are considered fake news. False information is disseminated by news organisations and social media in an effort to boost readership or wage psychological warfare. Our approach, false News Detection by Passive Aggressive Model, efficiently identifies fake news in texts extracted from social media platforms by utilising machine learning techniques. It is a computational stylistic study based on natural language processing (NLP). Fake news is categorised using the Passive Aggressive Model in a variety of ways. Based on the premise that real and fake news have distinct probability distributions when examining the module of their representation in the vector space of word frequency, the technique investigates the detection strategy for a statistical scenario. Our suggested passive-aggressive model outperformed the datasets in terms of recall and accuracy. Comparing our Passive Aggressive methods to other models, we obtain 93% accuracy.