Capstone Presentation

Alejandra Barrera
28/09/2020

Shinny App Basics

The basic idea with this project was to build a text prediction model using the users input. The key ideas are:

  • A predictive model was build using text from tweets, news and blogs.
  • Some n-grams models were used to build the model
  • The prediction was implemented through the shiny app platform

Data Exploration

First, some data exploration was made. It can be seen in the next table the main characteristics of the three sources

load("stat.RData")
knitr::kable(Stats)
Blogs_stats News_stats Tweets_stats
Lines 899288 77259 2360148
NumWords 37334131 2643969 30373543
maxnumwords_perline 6630 1031 47
minnumwords_perline 1 1 1

Some data visualization

Here I show some data visualizations. Performed in the exploratory data analysis. plot of chunk unnamed-chunk-2

Shiny app image

This is the main screen for the shiny app. It has a submit buttom were th input placed in the textbox is evaluated. The result appears next to the “Next word is”

plot of chunk unnamed-chunk-3

Thanks