I just got back from Strata + Hadoop World 2015 in London. I attended a lot of great sessions, but felt I wanted to be in two or three or four places at once. I got session envy overhearing attendees discussing sessions I had missed and reviewing the #StrataHadoop twitter stream.

The conference was 3 days long, with 10 tutorials and over 72 sessions. According to the site, in 3 weeks a link to a full set of recordings will appear in my email inbox. I wanted to know which sessions I should watch when the recordings become available. I also wanted to practice what I learned during the Spark Camp tutorial I attended with Paco Nathan. In a perfect convergence of problem meets tool, I installed Apache Spark and developed some ruby and pyspark scripts to pull #StrataHadoop and perform a simplistic text analysis to determine which talks were popular on twitter.

I also used this opportunity use RStudio’s Shiny web application framework and deployed the result via shinyapps.io. I highly recommend this combination. Using the tutorial, the entire effort of building and deploying on shinyapps.io was painless.

The code is on github: strata-hadoop-twitter-spark.

And here is a really awesome iframe of aimeeb.shinyapps.io/shiny-strata-titles: