Shiny RStudio’s Web Application Framework
Oftentimes clients want to benefit from recent advances in data science. They set aside their cynicism and skepticism and embrace the potential of predictive analytics. Clients want to evolve beyond data visualizations with dashboards and their concomitant drill downs, filters, and storytelling capabilities. With a bit of help from software marketers in the visual analytics space, Tableau, Qlik, Domo, etc, clients start to see predictive analytics as a silver bullet for the company’s strategic woes. What starts as an initial excitement for discovering some actionable insights can quickly devolve into a discussion of which is the best tool for the organization. The preponderance of an organization’s needs can and are solved by products in the parameterized reporting domain. In the predictive analytics domain, competitive pressures have contributed to a situation where all vendors are trying to up their game.
Shiny RStudio’s Web Application Framework
One tool, I have seen climbing up the value chain lately is RStudio’s Shiny. I come from a software development background so I am accustomed to web development frameworks. However, more and more folks are coming into the predictive analytics space from a data science background. They are used to tools like Matlab, r… For those folks, Shiny provides a nice web development framework to publish in html the results from our data science efforts on a back-end server. Here is a quick intro so you can judge for yourself.
Getting Started
To get started:
-
- In RStudio console:
-
- install.packages(shiny)
- library(shiny)
-
- Open a new file as a Shiny Web App
-
- Choose a name for your app
- Click on the Custom dialog box
This creates a small application wherein you can edit html, css, & javascript in the ui.R file. In the server.R file you can put your r code that your accustomed to.
Here is the sample code that is provided in the ui.R file:
library(shiny)
# Define UI for application that draws a histogram
shinyUI(fluidPage(
# Application title
titlePanel(“TaylorMade P790 Irons Velocity Test”),
# Sidebar with a slider input for number of bins
sidebarLayout(
sidebarPanel(
sliderInput(“bins”,
“Number of bins:”,
min = 1,
max = 50,
value = 30)
),
# Show a plot of the generated distribution
mainPanel(
plotOutput(“distPlot”)
)
)
))
We are drowning in information and starving for knowledge.
– Rutherford D. Roger
Next up: Shiny Gadgets
Robert Greenberg
Posted at 10:59h, 16 Maymanipulate was great, but now the world is demanding more utilities that you can get in shiny gadgets
Pavel Blatny
Posted at 10:59h, 16 MayI like to use manipulate but I think that is deprecated now.
Robert Greenberg
Posted at 11:01h, 16 Mayshiny rocks!!! You can also use shiny gadgets to do some exploratory data analysis