Applied Predictive Modeling in R

At useR! 2014 I'll be doing a morning tutorial (09:15 – 12:45) on at Monday, 30 June.

This course will provide an overview of using R for supervised learning. The session will step through the process of building, visualizing, testing and comparing models that are focused on prediction. The goal of the course is to provide a thorough workflow in R that can be used with many different modeling techniques. A case study is used to illustrate functionality. Basic familiarity with R is required.

Outline:

Introduction (philosophy, case study)
General strategies (data splitting, resampling, model tuning)
Data pre-processing (transformations, variable filtering)
Conventions in R (function interfaces, consistency)
Building and tuning models (performance metrics, trees, kernel methods)
Other Topics (as time allows) (comparing models via resampling, parallel processing)

I will be giving away about a dozen copies of the book (via names in a hat). Originally the book was to be included but since the tutorial is free, that isn't possible.

Materials

Slides
Code

Installing Packages:

Note to Windows users: The last two classes that I have taught where Windows users installed packages via RStudio, all the dependencies have not been installed. I recommend installing packages using base R.

install.packages(c("caret", "pROC", "rpart", "partykit", 
                   "C50", "kernlab", "AppliedPredictiveModeling",
                   "earth", "mda", "nnet"),
                dependencies = c("Depends", "Imports", "Suggests"))

You can test the install by loading caret via

library(caret)