EXPLORING THE EMOTIONAL FINGERPRINT OF FAKE NEWS: A COMPARATIVE SENTIMENT ANALYSIS OF TRUE AND FAKE NEWS ARTICLES

Abstract

Employing sentiment analysis to deduce the articles' emotional connotation and the Naive Bayes Classifier algorithm to differentiate between real and false news is the focus of this work. We evaluate the Naive Bayes Classifier's performance in correctly classifying news material using a dataset that includes labelled real and false news. We also use sentiment analysis to see how real news and false news compare emotionally and subjectively. Contributing to a better understanding of disinformation and its effects on public perception, this research seeks to shed light on the different sentiment patterns of real and fake news pieces. The results of this study have important consequences for the areas of media literacy, the identification of disinformation, and the creation of automated systems to counteract disinformation.