I've been thinking about how to build web applications that allow someone to do some data exploration interactively with some really nice visualizations. There are all kinds of choices such as Shiny for R and generic tools like AngularJS, D3.js, etc for general use. Of course there are also tools such as Tableau that provide their own interface. I'll probably try out Tableau public as well.

Those of you that know me are well aware that I don't build web front ends, so all of this is pretty foreign to me. So in an attempt to learn about the possibilities and get a chance to play with two of my favorite tools, R and Clojure, I'm going to try to build a web app.

Right now my plan is to try using the following tools in different ways:

As you can see I don't really know what I don't know at this point. The one I'm most familiar with is R/Shiny so that's what I'll be starting with.

Data

I'm going to use the data from Project Tycho at University of Pittsburgh. I'll be using the Level 2 data. The data consists of counts of 47 contagious diseases at the city and state level for the period 1888-2013. This seems like an interesting data set that shouldn't be too difficult to work with.

I'll use the first implementation in Shiny to explore what kind of visualizations and interactivity makes sense for this dataset. After that I'll attempt to implement the same thing using Clojure and some standard web app tools. From there I'll move on to things like d3.js and AngularJS.

So, keep an eye out here over the next few weeks as I explore how to build interactive web apps for data exploration and visualization. If you'd like to follow along with the code I'll be posting it up to my Github repository web-app-explore.



blog comments powered by Disqus

Published

15 January 2014