Identification of Fake News in Social Media Using Sentimental Analysis

This research paper focuses on finding the most effective models that promote detecting fake news from social media platforms. One of the ways to detect fake news is by text mining using sentiment analysis. In this paper, we will calculate the sentiment polarity using a lexicon-based approach and find out its role in fake news. In addition, we will compare the accuracy of Logistic Regression, Naive Bayes and Support Vector Machine which are supervised Machine Learning model for their performance in the detection of fake news in the Indian Fake New Detection dataset. As the result, we can see the role of sentimental score in detecting fake news and hyper tunning SVM model is the best model compared with the other supervised learning ML model, with accuracy of 93 %.