This app is designed to predict the next word in a sentence or phrase based on a word that the user enters.
When implemented on mobile devices this app offers improved typing speed and improved spelling/accuracy over standard text applications.
The Algorithm
This app uses an algorithm created using US Blogs, News, Twitter text data in English provided courtsey of SwiftKey and Coursera which determines word frequency.
First, the dataset is screened and processed to removed extraneous characters and then is categorized into the most frequent word combinations (N-grams).
Using these N-gram frequencies this app can take the user submitted sentences and quickly calculate the most likely next word.
App Interface
Below is the app interface- note it cannot take input in this presentation mode.
<!–html_preserve–>
The Next Word
You input the following text:
Your statement has been reformated to the following:
Most Likely Next Word:
The program guessed your word based on the following data:
<!–/html_preserve–>
Instructions for Use and Link to the App
The user enters his/her custo word or statement in the text entry box on the lefthand side
After clicking “predict the next word,” the app returns the most likely next word
Users can also view other likely next words and their corresponding probabilities on the bottom of the screen