Data Science Capstone Presentation

Creating a Predictive Text App

June 14 2020

This project is part of the Data Science Specialization Course offered by Johns Hopkins University on Coursera. Additionally, this project is done in collaboration with Swiftkey, a company that specializes in predictive text modeling.

The goal of this project was to use text analysis and natural language processing to create a predictive text model that can be accessed through an app hosted on Shiny.

The predictive text model was created using a Kat's back-off model.

I also consulted the following resources: 1 2 3 to create my text prediction model 4.

The Shiny app can be found here.

The Shiny app has a simple interface. To use it, one simply needs to enter the beginning of a sentence (e.g. I like to listen to the).

Then the user can click on the Predict Text button to generate the predicted text.

The full code used for this project (including this presentation) can be found here.

For more information on the Data Science Specialization course, visit Coursera.

I would like to end by thanking the instructors of the course. I gained new skills and knowledge that will be very useful in the future.

Good luck to everyone on your data science journey!