Abstract
We explore the use of machine learning techniques to classify a news source for generating unreliable news. Since the advent of the Internet, unreliable news and hoaxes have deceived users. Social media and news outlets are spreading false information to increase the number of viewers or as a part of the psychological competition. In this paper, we present an ensemble classifier using a set of marked true and bogus news articles. Here, the authors develop a classification approach based on text using SVM, Random-Forest, Naive Bayes, Decision Tree as a base learner in Bagging and AdaBoost. The purpose behind the work is to think of an answer that enable the user to classify and filter some of the false material. Accordingly, we show that the best performing classifiers were AdaBoost-LinearSVM and AdaBoost-Random Forest with 90.70% and 80.17% accuracy, respectively.