Data Science Capstone Predictive App Presentation

Konrad Zdeb
January 2016

Introduction

The following document summarised the development and functionalities of the word prediction application developed for the purposed of the Data Science Capstone offered via Coursera.

Presentation outline

Provided data
Analysis
Application preview
Summary / Links

Provided data files

The application is developed with use of the data files that were initially provided in the course. The characteristics of those files are given in the table below. Files summary

Predictive examples

The application makes use of the text analytics capacities available in R such as packages tm and RWeka to build matricies that can be used to suggest most likely words. The chart below illustrates example results of likely words accurding after the word data.

Sample

Application

The working application is presented in the screenshot below. Workinh application You can access the application at: https://konrad.shinyapps.io/CapstoneApp/