This project sheds light on a comparative study of different machine learning technique like Naïve Bayes (NB), Support Vector Machine, and Deep Neural Network used for text classification of Nepali documents which were scraped through news articles published from different online portals using their RSS (Really Simple Syndication) web feed.