Data Science Capstone Project

Objective

Develop a predictive text application capable of suggesting the next word in a sentence using Natural Language Processing (NLP) techniques.

Technologies Used

• R Programming Language
• Shiny
• Data.Table
• Stringr
• N-Gram Language Models

Business Value

• Faster text entry
• Improved user experience
• Demonstration of NLP capabilities in R

The application is available in the following link: https://roddymendozamarriott.shinyapps.io/final-project/

Prediction Algorithm

The algorithm searches for the most probable next word using:

• Quadgram Model
• Trigram Model
• Bigram Model

Example

Input sentence:

love

Best Prediction:

you

The text data were:

• Converted to lowercase
• Cleaned from punctuation
• Cleaned from numbers
• Stripped of extra spaces