In today’s digital era, proliferation of news or articles is a significant threat which affects integrity of data. Fake news detection is the process to recognized and verify accuracy of information whether news is false or misleading. In this research paper, comprehensive study of fake news detection, introduction about real or fake news, fake news detection taxonomy, and literature of various methodology of fake news detections are discussed. In this paper, machine learning model is developed which will be able to detect real or fake news. Here, three machine learning models used are LR model, RF, and decision tree model and performance evaluation of the model. Model is evaluated by using LIAR dataset which is a publicly available dataset. So, after performing the results, decision tree has given effective and efficient score as compared to the rest of the two machine learning classifiers. Decision tree has achieved the highest 100% result in terms of performance, i.e., accuracy to detect fake news as compared to another model. Multiple computational techniques are available that insights from linguistics and social network analysis. This proposed approach provides the effective solution for identifying the challenging task of detecting fake news in this digital era.