In the digitized world era, everything is made available to all people thus it leads to data dissemination and allows for sharing of unauthenticated information. There are several sources such as social media platforms, Twitter and web blogs that are sharing information. Here the need arises to predict whether the news is fake or real news. Fake news is essentially modeled as a distortion bias where true information is distorted by a degree to fit any biased view. The reason for the success and rapid spread of fake news is that everyone has inherent biases and looks for confirmation of their preexisting notions. It is used for furthering propaganda and sowing hate in society. It is therefore very important to develop systems to automatically detect fake news. In this paper, the Hard Voting Ensemble classifier is proposed for fake news detection. The use of Passive Aggressive Classifier, Decision Tree Classifier, Naive Bayes classifier, Support Vector Machine (SVM) and Voting Classifier on LIAR Dataset were studied in this paper. Feature engineering approach, TF-IDF vectorizer is used to transform the text into vector form. The performance of the proposed classifier is tested with various data sets.