ProjectWeek4 (Make a Shiny App)

Veronica Vaca
10/09/2018

Introduction

Build a Shiny App for the course Project

For this app I took as a reference this link in Rpubs https://rpubs.com/mohitagr18/239997, and I converted it in a reactive shiny app

We use the iris data set. The Petal's length and width and the Sepal's length and width to predict the class Species.

We can chose between 5 metodologies to compare the predictions

Pushing the Compute button will make the results to appear.

For running the app and these presentation without errors you have to install the following packages:

  • shiny, webshot,
  • caret,
  • ellipse,
  • e1071,
  • MASS,
  • plotly.

Data treatment

We are going to see some statistics of the data set and then split it into the training and testing set.

require(caret); require(plotly);require(shiny)
summary(iris)

  Sepal.Length    Sepal.Width     Petal.Length    Petal.Width   
 Min.   :4.300   Min.   :2.000   Min.   :1.000   Min.   :0.100  
 1st Qu.:5.100   1st Qu.:2.800   1st Qu.:1.600   1st Qu.:0.300  
 Median :5.800   Median :3.000   Median :4.350   Median :1.300  
 Mean   :5.843   Mean   :3.057   Mean   :3.758   Mean   :1.199  
 3rd Qu.:6.400   3rd Qu.:3.300   3rd Qu.:5.100   3rd Qu.:1.800  
 Max.   :7.900   Max.   :4.400   Max.   :6.900   Max.   :2.500  
       Species  
 setosa    :50  
 versicolor:50  
 virginica :50  




sample <- createDataPartition(iris$Species, p=0.80, list=FALSE)

iris_train <- iris[sample,]

Univariate Plots

Now that we have the training set we can make some plots of it, in the shiny app you can chose between a boxplot and a barplot, here we are going to see the boxplots. I couldn't put a plotly plot in the presentation but in the app the boxplots are made in Plotly.

plot of chunk unnamed-chunk-2

Multivariate Plots

For th multivariate plots in the shiny app you have 3 choices, scatter, box and density plots from the caret package.

You have to see this working on the following link. https://vero.shinyapps.io/appweek4/

And the code you can find in: https://github.com/vefra/DevelopingDataProductsProject

Here we have an example.

plot of chunk unnamed-chunk-3

Finally we make the comparisson between the models.