COVID FAKE NEWS CLASSIFICATION WITH NATURAL LANGUAGE PROCESSING
Raj Sanjay Kulkarni, S Vinod, Beatriz Lúcia Salvador Bizotto, Maria Emília Camargo, Mithileysh Sathiyanarayanan · Zenodo (CERN European Organization for Nuclear Research) · 2022
we are already dealing with a pandemic of fake news in today's rapidly expanding world of technology and journalism. It is difficult to predict if a news story is fake or authentic because there are thousands of sites on the internet which suggest different things. This type of misleading information causes a lot of violence and chaos among people. To avoid this, we have worked on ―COVID Fake News Classification‖. The main aim of this paper is to develop a classifier that segregates the news into fake and authentic with an accuracy around 90%. We collected a public dataset consisting of news and used simple NLP technique for the pilot study; compared and validated our trained model using three machine learning techniques – Multinomial Naive Bayes, Passive Aggressive and Logistic Regression (LR) where Logistic Regression showed the maximum accuracy (~93%).