Online News Article Classification Using Machine Learning Approaches

Multiple forms of information are fed into the internet news portals from numerous sources. In the majority of real-life settings, it is highly desired to categorize this information into the proper groups, and it is crucial to have a reliable mechanism in place for dividing news into these groups. The classification system is strengthened and improved using machine learning because there has been little research done on how to classify news headlines, there is now a chance to examine this subject in more detail. This work focuses on categorizing real-time news based on its headlines. A system has been designed to classify each news headline to its pre-defined category. The model is trained such that the machine can predict the category of the news item accurately. The news headline will be fetched in real time and will be passed through this classifier. This whole process will not only lead to a better working model but also show a comparative study of naïve bayes and logistic regression models for classifying news headline. Logistic Regression produces the best outcomes compared to Naïve Bayes.