The goal of this project is to take a dataset provided and create an NLP (natural language processing) model that is able to predict subsequent words. Blogs, Twitter and News were datasets used to train the model. SwiftKey is the company that works in cooperation with Professors of the Johns Hopkins University to prepare this Project, with objective to construct a predictive model that makes easier for people to type on their mobile devices. Besides cleaning and sub-setting the data, the tokenization technique of N-Grams were used to combinations the words to be used at the predictive algotithm. The final project was concluded with one Shiny application and a Pitch using R-Presentation.