Johns Hopkins Data Science Capstone

Author: Yash Gautam

Objective

Develop a predictive text application that predicts the next word using n-gram language models.

Data Sources

• Blogs
• News
• Twitter

Dataset Summary

• More than 4 million text entries
• Over 100 million words
• English language dataset
• Used for predictive text modeling

Method

• Convert text to lowercase
• Remove punctuation
• Remove numbers
• Remove extra spaces
• Build unigram, bigram and trigram models
• Use a backoff strategy to predict the next word

Features

• Enter a phrase in the text box
• Click Predict Next Word
• Displays the predicted next word

Live Application

https://yashgautam.shinyapps.io/SwiftKeyPredictor/

Summary

• Interactive next-word prediction application
• Built using R and Shiny
• Successfully deployed online

Future Improvements

• Improve prediction accuracy
• Use larger datasets
• Support multiple suggestions

Thank You