Abstract Details
Activity Number:
|
79
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
|
Sponsor:
|
Section on Statistical Graphics
|
Abstract #311588
|
|
Title:
|
Glassbox: An R Package for Visualizing Algorithmic Models
|
Author(s):
|
Max Ghenis*+ and Ben Ogorek and Estevan Flores
|
Companies:
|
Google and Google and Google
|
Keywords:
|
R ;
Random forests ;
Data visualization ;
Prediction ;
ggplot2 ;
Machine learning
|
Abstract:
|
Black-box prediction models like random forests often strike an ideal balance between robustness, accuracy, and ease of use for analysts. In a business setting, however, the difficulty of interpretation is a drawback.
We propose a novel approach for visualizing such models using model predictions across predictors' observed ranges. By smoothing the predicted trend of the test set's response curves, we detect meaningful relationships which would otherwise remain masked.
We introduce the Glassbox R package which implements these techniques for any prediction model, and describe examples of Glassbox's usage from our studies in Google's People Operations department.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.