7月 272016
 

Bellagio Hotel for Analytics ExperienceFirst it was the patriarch of my favorite family. That shocking Red Wedding scene meant I could cross off several more characters I’d grown to love. When the season finale of Season 5 left me asking if we’d lost yet another one of my favorites, I wasn’t sure how much more I could take. Of course, I’m talking about the often surprising deaths of some of our favorite characters in HBO’s wildly popular series Game of Thrones. If you’re a fan of the show you know that no one, regardless of how important they are to the storyline, is safe from an untimely demise.

Though I’ve grown somewhat accustomed to saying goodbye, it sure would be nice to know the likelihood a particular character will live or die, just so I can prepare for the heartache in advance if need be. Thankfully, Taylor Larkin, a student at the University of Alabama, thinks survival data mining can help. He plans to show us how in an e-poster he'll present at this year’s Analytics Experience conference, September 12 – 14, at the Bellagio hotel in Las Vegas.

Using plot points from the books the TV series is based on, along with survival data mining using the Survival node in SAS® Enterprise Miner™ 13.1, Larkin has created an analysis that estimates the probabilities popular Game of Thrones characters will survive through time. His research was inspired by the analysis and datasets created by Olin College Computer Science Professor Allen Downey and some of his students, who used Bayesian survival analysis to do something similar.

Analytics Experience conference presentations

Larkin’s sure-to-be awesome presentation is one I’m really looking forward to seeing, but it’s also just one of more than 100 talks planned for the event. The Analytics Experience conference provides attendees an in-depth look at some of the latest research, top trends and new techniques being used in the field of analytics. This is the nineteenth consecutive year that SAS has hosted the event.

This year’s conference offers six keynote addresses and dozens of session talks. Some of the topics presenters will cover include customer intelligence, business intelligence, data management, Hadoop, fraud, cybersecurity, risk analytics, and the Internet of Things. E-Poster presentations, demos, training classes and table talks provide even more insight and allow attendees to explore other creative ways to use analytics.

A number of the talks will come from university students – I always find them to be a great addition to the conference’s content. Besides Larkin’s talk, student presenters will show you how to use analytics to do things like detect sarcasm on social media, build a restaurant recommender engine or defend Steph Curry, something the rest of the NBA couldn’t seem to do last season.

Though talks are still being added every day, a large portion of conference presentations are now available.

Keynote speakers include:

  • Jared Cohen, President of Jigsaw and Chief Advisor to the Executive Chairman of Alphabet.
  • Jim Goodnight, CEO of SAS.
  • Amber MacArthur, President of Konnekt.
  • Jeremiah Owyang, Founder of Crowd Companies.
  • Jake Porway, Founder and Executive Director of DataKind.
  • R. Ray Wang, Principal Analyst, Founder and Chairman of Constellation Research.

If you’re in the field of analytics, there isn’t a better conference to help advance your knowledge. It’s an event I look forward to every year and I hope to see you there.

For more information, visit the website or join the community dedicated to the event. You can also view a number of videos from last year’s event on YouTube.

P.S. If you do make it to Vegas in September, let’s plan to meet at Larkin’s e-poster. We’ll find out together how likely Tyrion is to make it safely through Season 6!

tags: analytics, analytics conference, analytics experience, Game of Thrones

Will your favorite Game of Thrones character survive next season? was published on SAS Users.

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

(required)

(required)