Crowdsourcing Data Science

Our main speaker this month is Anthony Goldbloom, the founder and CEO of Kaggle (http://www.kaggle.com), who will talk about how R is used for the prediction competitions. We'll also have two lightning talks from Norm Matloff and Joseph Rickert.

Please Note: The Giants will be playing at nearby AT&T Park, so parking will be limited. Thanks to Manish from RapLeaf for finding our new venue.

Agenda:

6:30 Networking and refreshments (with pizza provided by Dataspora and Rapleaf)

• With two copies of "R Cookbook (http://oreilly.com/catalog/9780596809164)" to give away, thanks to O'Reilly.

7:00 Announcements and introductions

7:10 "Statistical Software Alchemy: Transforming Complex Problems into Embarrassingly Parallel Ones" - Norm Matloff, Professor of Computer Science, UC Davis

7:20 "Extracting Time Series from Large Data Sets" - Joseph B. Rickert, Technical Marketing Manager, Revolution Analytics

7:30 "Crowdsourcing Data Science" - Anthony Goldbloom, Founder and CEO, Kaggle.

8:30 Close

Crowdsourcing data science

Predictive modeling competitions are shaping up to be the biggest thing in data science in 2011 - there are multi-million dollar prizes on the way and vital scientific projects are being thrown open to competition for the first time. Approximately 25 percent of the models submitted in Kaggle competitions so far have been based on R, the open source statistical modeling and programming language. Anthony Goldbloom, Kaggle’s founder and CEO will describe the ideas underlying the competitions and discuss the use of R in building predictive models. Anthony will also discuss a past competition ( http://www.kaggle.com/R ) to build a recommendation engine for R libraries as well as the $3 million Heritage Health Prize.