The data set consist of ~ 3 million lines of English text from various news articles, twitter posts and blogs. This application uses n-gram model to predict next word.
Dataset Download Link: Here
nextword1.R : Cleans the data + extract n-gram + save term document matrix nextword2.R : Uses term document matrix and predicts next word
The Shiny application: https://sarthakdasadia.shinyapps.io/NextWord