Capstone Project

Nashit Babber
Sept 2, 2017

Overview

This capstone project was given with the following goals:

To develop a shiny app that could predict the nest word
creating a presentation to get a pitch from the investors

The shiny app can be found at the following URL

Approach and Logic

First the data was loaded and sanitised.
Text were tokenised into n-grams for prediction.
The final n-gram data frames are the matrix of words with their frequencies to predict the text based on user input.
if the word cannot be predicted it just prints null.(Try some words like hakuna matata ;))

The Shiny App

my image

Start typing in the input box(in the side panel) , the predicted word will appear in the main panel.The model used to predict the word will be displayed below the predicted word

Wrap Up

This model can be made more accurate, but this could have led in serious drop off in the speed and would hae increased the computation power.
The Shiny app for this project could be found at this link
This pitch can be found at this Rpubs URL