Online Fake News Prediction on Social Media Using Machine Learning

Basically recommendation system generates based on profiles of users news benefits based on their past historical browsing behavior for such users who connected with the system recently as well as explicitly allowed web history. To produce personalized news recommendations, combine the information filtering mechanism with the user profiles experienced with the current collaborative filtering mechanism. To build a customized news recommendation system, use the popular micro blogging service using Facebook. The proposed research provides online news recommendation using hybrid machine learning algorithm. System initially deals with Natural language Processing (NLP) to extract the features and train the module respectively. The system can recommend the news based on user personalized history, various dataset have been evaluate to measure the performance analysis of system which provides better prediction accuracy than other recommendation systems.