Instead of using traditional newspapers as much to disseminate news, news companies are now using digital media. While sensitive material that is inappropriate for a certain age group may be present, a significant quantity of information that is currently saved in electronic format needs to be divided into several categories. In this research, a dataset is utilized to classify news employing machine learning methods. We will choose the algorithm that offers the greatest degree of precision by comparing and contrasting the performance of the following techniques: decision tree, logistic regression, linear regression, and Naive Bayes. Additionally, news pieces can be proposed to online readers depending on their interest piqued by past reads and reader feedback, or suggested depending on how closely a news article matches the news users are now reading. Following implementation and accuracy comparison across all Machine Learning models, the Naive Bayes Model produced the best results with the lowest error percentage. Thus, the Naive Bayes Model is employed to train the model and obtain the desired outcome.