Introduction to R for Data Mining

Joseph Rickert (Revolution Analytics)

9:00am Tuesday, 02/28/2012

Data Science, Ballroom G

Please note: to attend, your registration must include Tutorials.

Presentation: Introduction to R for Data Mining Presentation [PPTX]

Average rating:

(4.50, 4 ratings)

Attendees: We will be using Revolution R Enterprise 5.0.1 during the tutorial. Download a 90-day evaluation copy of Revolution R Enterprise software as well as the scripts and other auxiliary material. The file STRATA2012scripts.zip contains 40 R scripts that form the basis of the tutorial. The auxiliary material is in two files:

pdfs.zip contains some optional reference material
movies.zip contains a data used in a data base example. It is not essential for the tutorial but is included for completeness.

The R scripts were tested on R 2.13.2. However, there should be no problem running them on R 2.14.1. The scripts also require a number of packages to be loaded. The primary package is rattle. When rattle loads, it also loads most of the other packages that will be needed. Additionally, near the top of each script required packages will be identified by line of code similar to library(ggplot2). CRAN packages can be dound and downloaded here.

Only 5 of the 40 scripts require Revolution R Enterprise. All of the others can be run from open source R which you may download directly from CRAN.

This tutorial will enable anyone with some programming experience to begin analyzing data with the R programming language

Syllabus

Where did R come from?
What makes R different from other statistical software?
Data structures in R
- Reading and writing data sets
- Manipulating Data
Basic statistics in R
- Exploratory Data Analysis
- Multiple Regression
- Logistic Regression
Data mining in R
- Cluster analysis
- Classification algorithms
Working with Big Data
- Challenges
- Extensions to R for big data
Where to go from here?
- The R community
- Resources for learning R
- Getting help

People planning to attend this session also want to see:

Joseph Rickert

Revolution Analytics

I am a marketing manager at Revolution Analytics with a passion for analyzing data. I have worked a number of successful Silicon Valley start-ups including Sytek, Alantec, Parallan Computer and Scotts-Valley Instruments. I have graduate degrees in both the Humanities and Statistics. I taught statistics briefly at SJSU and I blog at blog.revolutionanalytics.com

Comments on this page are now closed.

Comments

Bharath Mundlapudi

03/17/2012 1:41pm PDT

Hi, How i can i download this file STRATA2012scripts.zip? Please advice.

Sophia DeMartini

02/29/2012 10:56am PST

Hi Michael,

I'm sorry, the information I provided was actually incorrect.

This is actually called “Complete Video Compilation” and it is not available on oreilly.com. It can be pre-ordered at the following URL:

strataconf.com/strata2012/p...

Please let me know if you have any other questions. I can also be reached via email at sophia@oreilly.com.

Thank you, Sophia

Sophia DeMartini

02/29/2012 10:13am PST

Hi Michael,

We're recording all tutorials and sessions at Strata, and they'll be available after the conference as part of the All Access video compilation, which can be purchased on oreilly.com.

Best, Sophia

Micheál Keane

02/28/2012 6:23pm PST

Couldn’t make this because I was at the Deep Data series. Will a video of this be posted?

Zebulon Young

02/28/2012 4:16pm PST

Nathan: The ZIP files identified above are each available by following the link at the top of the instructions (“Download a 90-day evaluation copy of Revolution R Enterprise”), after completing a registration form. I think that since they are all behind this registration form, it wasn’t appropriate to link to them each directly. Hope this information helps.

Nathan Wenzel

02/28/2012 4:09pm PST

I see the names of the code zip files, but they don’t apear to be links. Where can the .R code from the presentation be found?

Devender Gollapally

02/27/2012 4:03pm PST

R studio seems nice

Joseph Rickert

02/27/2012 8:31am PST

Sorry for the late down load notice. (1) Revolution R Enterprise is not required for running most of the scripts. I will use it to show R handling a fairly large data set.

(2) The other scripts will run on standard CRAN R

(3) Revolution R Enterprise will not run on a MAC

(4) Rattle would be nice to have loaded if you can. I am running it with Revolution R Enterprise that has R 2.13.2 underneath

Leigh Dodds

02/27/2012 7:28am PST

If anyone else on Ubuntu has Rattle installation issues, then the steps here worked for me:

groups.google.com/group/rat... users/msg/c6cff79f3960295b

Leigh Dodds

02/27/2012 7:22am PST

I only have Ubuntu available so can’t use Revolution. The community edition which does run on Ubuntu seems to be too old.

I’ve also encountered GTK+ issues when trying to install Rattle. This seems to be a common issue with the latest version.

Are there other options?

Matthew Feinberg

02/27/2012 6:59am PST

I’m in a similar boat to Richard. Just found out about the Windows/RHEL5 requirement when all I have is Ubuntu and OSX. Hopefully there won’t be much focus on Revolution specific features, because there’s no time to work around this requirement.