Predictive Analytics World is hosting our March Meetup
Agenda: 5:30 - 7:00 - PAW Reception
7:00 - 7:15 - Greeting from Eric Siegel
7:15 - 7:20 - Announcements
7:20 - 7:50 - Max Kuhn- Predictive Models in R Clustered by Tag Similarity
8:00 - 8:30 - Antonio Piccolboni - plyrmr: Hadoop Mapreduce for the rest of us
8:30 - 9:00 - Megan Price - Statistics, R and Human Rights
Note: You must RSVP to attend this meeting. Members who are on the meeting list are invited to attend the PAW reception which will be held in the Exhibit Hall
Max is Director of Nonclinical Statistics at Pfizer and author of R's caret package. He will talk about his work to refactor the caret package focusing on his method of classifying the predictive models included in caret and visualizing their relationships.
Antonio is a data scientist, researcher and the major force behind RHadoop. He writes:
The `rmr` package made it possible to write Hadoop mapreduce programs in R and some users estimated the productivity gain to be roughly 30X! But that wasn't enough and we were asked if it was possible to make it easier. Thus `plyrmr` was conceived. It's the latest addition to the RHadoop open source project, supported by Revolution Analytics. It takes inspiration from the highly popular data manipulation package `plyr` and SQL to create a kind of specialized language within R that should look familiar to most R users, yet it is backed by the power of Hadoop. To get a feel for it, this is a snippet written with `plyrmr`:
select(model = list(lm(mpg~cyl+disp))) %|%
Can you guess what it does? This talk will explain how to write many more programs like this.
As the director of research at the Human Rights Data Analysis Group, Megan Price designs strategies and methods for statistical analysis of human rights data for projects in a variety of locations including Guatemala, Colombia, and Syria.