Accurate Fake News Prediction by Comparing Performance of Machine learning algorithms

Abstract

With the advancement of technology and the widespread use of media there has been an increase, in the circulation of fake news. Unfortunately, some individuals intentionally spread information to manipulate opinion and drive traffic to specific websites. One such instance occurred during the Covid 19 pandemic when misleading rumors started circulating falsely claiming that Covid vaccines were linked to heart attacks and infertility. These baseless claims created hesitancy among people regarding vaccination. To assist individuals in identifying news accurately this paper compares the performance of various machine learning algorithms such as Passive Aggressive Classifier, Decision Tree, Random Forest, Logistic Regression and Naïve Bayes. After evaluating their results, it was determined that the Passive Aggressive Classifier achieved an accuracy rate of 98.2% followed by Naïve Bayes with 96.59% accuracy Random Forest with 96.95% accuracy, Decision Tree with 96.23% accuracy and Logistic Regression with 97.22% accuracy. Based on these findings it can be concluded that the Passive Aggressive Classifier is the algorithm for predicting fake news among all five models tested in this study. The data used for building these machine learning models was obtained from Kaggle website. The primary objective of this research paper is to provide guidance to individuals seeking to choose an algorithm that offers accuracy, in detecting news.