Data Science Capstone final project

• This presentation is a brief introduction to the application which can be used to predict the next word based on the words you entered.
  
• This is application is the final project of coursera course Data Science Capstone.
  

This application used three text files as training data: blogs, news, and twitter: https://d396qusza40orc.cloudfront.net/dsscapstone/dataset/Coursera-SwiftKey.zip

• N-gram model is used for prediction of the next word based on previous 1, 2, or 3 words.
  
• The three text files were cleaned by removing numbers, removing punctuation, removing symbols, removing separators, and convert to lower case. Then the cleaned texts were tokenized into n-grams (unigram, bigram, trigram, and fourgram).
  
• The tokenized n-grams then sorted and saved as data frames in R. The data frames were used by the application for next word prediction.

• The application is hosted on shinyapps.io: https://mary0523.shinyapps.io/CapstoneProject/

• The code for this application and related files are stored in GitHub: https://github.com/Mary0523/DataScienceCapstoneProject