Sciweavers

Share
ICANN
2011
Springer

Bias of Importance Measures for Multi-valued Attributes and Solutions

10 years 3 months ago
Bias of Importance Measures for Multi-valued Attributes and Solutions
Attribute importance measures for supervised learning are important for improving both learning accuracy and interpretability. However, it is well-known there could be bias when the predictor attributes have different numbers of values. We propose two methods to solve the bias problem. One uses an out-of-bag sampling method called OOBForest and one, based on the new concept of a partial permutation test, is called pForest. The existing research has considered the bias problem only among irrelevant attributes and equally informative attributes, while we compare to existing methods in a situation where unequally informative attributes (with or without interactions) and irrelevant attributes co-exist. We observe that the existing methods are not always reliable for multi-valued predictors, while the proposed methods compare favorably in our experiments.
Houtao Deng, George C. Runger, Eugene Tuv
Added 29 Aug 2011
Updated 29 Aug 2011
Type Journal
Year 2011
Where ICANN
Authors Houtao Deng, George C. Runger, Eugene Tuv
Comments (0)
books