This presentation is to briefly pitch my application for predicting the next word.
The application is the capstone project for the Coursera Data Science specialization the Johns Hopkins University in cooperation with SwiftKey.
June 16, 2017
This presentation is to briefly pitch my application for predicting the next word.
The application is the capstone project for the Coursera Data Science specialization the Johns Hopkins University in cooperation with SwiftKey.
The goal of this project was to make a Shiny application that predicts the next word.
The idea of this project (and whole course) was to teach the students some data science concepts and methods.
In this specific application we were supposed to learn about natural language processing.
The data were taken from the provided corpora: news, blogs, and Twitter.
After taking a sample from the corpora it was cleaned in the following way:
The preprocessed sample was tokeniezed into n-grams (where n = 1..4.) The results are used for predicting the next word as a function of user input and the frequencies in the n-gram tables.
The user interface is kept extremely simple. There is one input for the user to write the text. Below the user is presented with a text box showing 3 most probable next words.
That's as easy as that : )
Data Science Specialization: https://www.coursera.org/specializations/jhu-data-science
The application: https://kelnerrr.shinyapps.io/dss_capstone_project/
The code: https://github.com/kelnerrr/DSS_Capstone
This pitch presentation: http://rpubs.com/kelnerrr/DSS_Capstone_Pitch