Coursera Data Science: Final Project

This is the slide deck for the final project of the Coursera Data Science Specialization Capstone course. The project is to build a predictive model using NLP to recommend the most likely next word given a string input.

The model itself was built using R and developed into a Shiny Application.

The Model was trained using data from SwiftKey and is from blogs, news, and tweets. The data is based in English.

The data can be found here: Data

The prediction application process begins with building corresponding n-grams from a sample of the data, and storing this data to save computational time. Using the N-gram data, we can predict the next word based on occurrence frequency.

The application can be found here:

Text Prediction Application

The Github Repository and source code for the app can be found here:

Git Repo