JOURNAL BROWSE
Search
Advanced SearchSearch Tips
Input Variable Importance in Supervised Learning Models
facebook(new window)  Pirnt(new window) E-mail(new window) Excel Download
 Title & Authors
Input Variable Importance in Supervised Learning Models
Huh, Myung-Hoe; Lee, Yong Goo;
  PDF(new window)
 Abstract
Statisticians, or data miners, are often requested to assess the importances of input variables in the given supervised learning model. For the purpose, one may rely on separate ad hoc measures depending on modeling types, such as linear regressions, the neural networks or trees. Consequently, the conceptual consistency in input variable importance measures is lacking, so that the measures cannot be directly used in comparing different types of models, which is often done in data mining processes, In this short communication, we propose a unified approach to the importance measurement of input variables. Our method uses sensitivity analysis which begins by perturbing the values of input variables and monitors the output change. Research scope is limited to the models for continuous output, although it is not difficult to extend the method to supervised learning models for categorical outcomes.
 Keywords
Supervised Learning;Input Variable Importance;Linear Regression;Neural Network;Regression Tree;Sensitivity Analysis;Data Mining;
 Language
English
 Cited by
1.
Bias Reduction in Split Variable Selection in C4.5,;;;

Communications for Statistical Applications and Methods, 2003. vol.10. 3, pp.627-635 crossref(new window)
1.
Bias Reduction in Split Variable Selection in C4.5, Communications for Statistical Applications and Methods, 2003, 10, 3, 627  crossref(new windwow)
 References
1.
Classification and Regression Trees, 1984.

2.
Korean Communications in Statistics, 2002. vol.9. pp.155-166 crossref(new window)

3.
Korean Journal of Applied Statistics, 2002. vol.15. pp.190-201 crossref(new window)

4.
The Elements of Statistical Learning, 2001.

5.
Korean Communications in Statistics, 1997. vol.4. pp.827-832

6.
Journal of Data Science and Classification (Korean Classification Society), 1997. vol.1. pp.69-84

7.
Korean Communications in Statistics, 2000. vol.7. pp.937-943

8.
Journal of Korean Statistical Society, 1996. vol.25. pp.1-24

9.
Korean Journal of Applied Statistics, 2001. vol.14. pp.91-101

10.
Pattern Recognition and Neural Network, 1996.

11.
How to measure importance of inputs? Unpublished White Paper, 1998.

12.
Korean Journal of Applied Statistics, 2001. vol.14. pp.475-486

13.
Clementine 7.0 User's Guide, 2002. pp.525

14.
Unpublished White Paper, 1997.

15.
Korean Communications in Statistics, 2002. vol.9. pp.809-824 crossref(new window)