UseR! 2013: it’s a wrap!

[This article was first published on Revolutions, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Steve Scott
Steve Scott (Google) presents at useR! 2013, July 12 2013

The 2013 UseR! conference has drawn to a close in Albacete, Spain. The conference organizers did a fantastic job putting together a jam-packed presentation and social program for the 350+ R users in attendance. Here are just a few of my highlights from the last couple of days:

Duncan Murdoch's keynote What's New in R 3.0.x also gave an early surprise about R 3.1: it will include performance improvements from Radford Neal's pqR. R 3.1 is scheduled for March 2014.

Kamson Lai revealed that GroupOn uses R to demonstrate to merchants that a sizeable proportion of coupon redeemers will probably become repeat customers. They use linear regression, MARS (from the earth package) and SVMs (from the e1071 package) to model 28 million transactions from 200k coupons.

Jose Benitez's keynote described the RSNNS package for “Soft Computing” based on fuzzy logic and fuzzy rules. 

Joe Cheng from RStudio showed how to create interactive Web apps with Shiny, and demonstrated an exciting (but as yet unreleased) “reactivity” package for R (where you can change an R variable, and have expressions, tables and charts based on that variable update instantly).

Tobias Verbeke from OpenAnalytics gave a live demonstration of the R Service Bus: send an email with an attached dataset, get an R-based analysis in reply. He also mentioned Architect, an Eclipse-based IDE for R. 

Steve Scott from Google (pictured) noted that his “BOOM” C++ library for Bayesian statistics provides the intelligence behind some of Google's user-facing apps. (He plans to release an R package interface to CRAN soon.) He also described the problem of “Nowcasting“, for which Google search query frequency data provides useful predictors.  

Drew Schmidt shared his experiences running R on a supercomputer with 112,000 cores and 150 Terabytes of RAM, with his pbdR package. (The Programming with Big Data in R vignette provides detailed background info.)

Hannes Mühleisen showed some impressive benchmarks using the MonetDB.R package, which provides a drop-in replacement for data frames where the data resides in the MonetDB column store instead of in memory.

Again, that's just a few highlights — there were so many fascinating talks, and I could only get to a fraction of them. Kudos to the organizing committee: we're very proud here at Revolution Analytics to have been sponsors of such a wonderful event. I'm already looking forward to next year's conference, useR! 2014, to be held in Los Angeles June 30 – July 3. 

To leave a comment for the author, please follow the link and comment on their blog: Revolutions.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)