A CASE STUDY ON FAKE NEWS DETECTION USING MACHINE LEARNING ALGORITHMS

Abstract

The vast majority of the world's population reads or analyses news produced on the internet, whether via social media or elsewhere. The news is disseminated over the internet without the need to check its veracity. In this work, we experimented with a system to categorise the news in the dataset in order to determine if the news is real or false. We combined the Count Vectorizer with the MultinomialNB (multinomial Naive Bayes classifier) and the term frequency-inverse document frequency (TF-IDF) vectorizer with the PassiveAggressiveClassifier. A dataset called "fake or real news.csv" was used to train and test the algorithm. The output of the system reflects its efficiency.