Performance Comparison of Machine Learning Classifiers for Fake News Detection

Information sharing on the web particularly via web-based networking media is increasing. Ability to identify, evaluate and address such information is significantly important. Fake information deliberately created is purposefully or unintentionally engendered over the internet. This is affecting a larger group of society who are blinded by technology. This paper illustrates model and methodology to detect fake news from news article with the assistance of Machine learning and Natural language processing. In this proposed work different feature engineering methods like count vector, TF-IDF and word embedding are used to generate feature vector. Seven different Machine learning Classification algorithms are trained to classify news as fake or real and are compared considering accuracy, F1 Score, recall, precision and best one is selected to build a model to classify news as fake or real.