Intervention in prediction measure: a new approach to assessing variable importance for random forests

Intervention in prediction measure: a new approach to assessing variable importance for random forests

Abstract Background Random forests are a popular method in many fields since they can be successfully applied to complex data, with a small sample size, complex interactions and correlations, mixed type predictors, etc. Furthermore, they provide variable importance measures that aid qualitative inte...

Full description

Bibliographic Details
Main Author: Irene Epifanio
Format: Article
Language:English
Published: BioMed Central 2017-05-01
Series:BMC Bioinformatics
Subjects:
Random forest
Variable importance measure
Multivariate response
Feature selection
Conditional inference trees
Online Access:http://link.springer.com/article/10.1186/s12859-017-1650-8

Internet

http://link.springer.com/article/10.1186/s12859-017-1650-8

Similar Items