To reiterate, the goal of this project was to create a machine learning model using NLP techniques that will accurately detect whether a news article is true or fake. Using Naive Bayes, the model created with count vectorized text data performed the best with an accuracy of 90.6%. Using Support Vector Machines the model created with the count vectorized text data using a polynomial kernel with a degree of 4 performed the best with an accuracy of 98%. The best model created using the term-frequency vectorized text data was using the rbf (gaussian) kernel, with an accuracy of 96.8%. In general, the Support Vector Machines Models performed better than the Naive Bayes models. The creation of a model with a 98% accuracy is considered a great success, as providing people with a 98% confidence that the article they are reading is true or fake can only be beneficial. It will allow people to make more informed decisions and form more accurate opinions based on the news they read.