This is my solution to the capstone project of the Coursera Data Science Specialization and it brings together the various subjects studied in the course - R Programming, data cleaning, data exploration, statistical inference, predictive model building and data product building
The aim of the project was to take build a predictive typing model that would suggest the next word to a user as they type. A statistical analysis of several large sets of publically available text data: blog posts; news articles; and Twitter posts, was performed by counting the frequency of 1,2,3 and 4 word ngrams. These frequencies were then used to build a model of the likelihoods of a particular given the three prior words. To show the model in operation,a Shiny-based text editor was built that allows the user to type text and that generates a suggested next word every time the user completes a word.