This is the final project submission for the Data Science Capstone Project which is part of Data Science track in Coursera.

The purpose of this project to build a prediction model to predict the next word given a sentence from the input.

• In order to build the model, a corpus was built by using the dataset which contains blog, news and twitter data.

• Sample size of 1% used in order to select from corpus so that performance of the model is manageable.

• After cleaning and sampling the data ngrams were built which contains unigrams, bigrams and trigrams.

• Model takes the last 2 words of the inputted sentence and uses the trigrams to predict the next word.

User interface for this application expects the user to enter a sentence and click submit.

A shiny application to test the model can be found in the following website: https://msoysal.shinyapps.io/Predict_Next_Word/

All the files for this project can be found in the following github page: https://github.com/marino35/DSCP