Want to know the predicted next word? Use our new app…

• Users start to type a sentence, the app predicts next word
• App uses a subset of data from the three data sources (blogs, twitter & news)
• App also uses the technology of Swiftkey

Try it now…select it here

• Click on the link for the app.
• Select the news article or twitter data for prediction.
• Type the word in “Enter text..” section.
• Use clean and clear option below to check the next word predicted based on datasets.

• Merged data from the 3 Data Sources into one data file (Blogs, Twitter & News)
• Cleansed data including converting to lowercase, stripping white space, and removing punctuation & numbers
• Created Bigram, Trigram and Quadgram n-grams
• Extracted term-count tables from the n-grams
• Sorted in descending order based on frequency
• Saved n-gram objects

• Algoithm checks for the highest-order n-gram (n=4)
  

• If n=4 is not found, then checks the next lower-order model (n=3)

• If n=3 is not found, then the app continues to check (n=2)

• If n=2 is not found, then the app returns “No Match Found”

• Code is available on GitHub

• Further work can include expanding both the number of data sources & number of n-grams Enjoy the app!