Random forest using tree selection method to classify unbalanced data

Baoxun Xu; Yunming Ye; Qiang Wang; Junjie Li; Xiaojun Chen

doi:10.1117/12.970545

8 June 2012 Random forest using tree selection method to classify unbalanced data

Baoxun Xu, Yunming Ye, Qiang Wang, Junjie Li, Xiaojun Chen

Proceedings Volume 8334, Fourth International Conference on Digital Image Processing (ICDIP 2012); 83344F (2012) https://doi.org/10.1117/12.970545
Event: Fourth International Conference on Digital Image Processing (ICDIP 2012), 2012, Kuala Lumpur, Malaysia

Abstract

Random forest is a popular classification algorithm used to build ensemble models of decision tree classifiers. However, owing to the complexity of unbalanced data distribution in high dimensional space, a random forest may include bad trees that can result in wrong results. This paper proposed an improved random forest algorithm with tree selection methods. This algorithm is particularly designed for analyzing unbalanced data. The novel tree selection methods are developed for making random forest framework well suited to classify unbalanced data. Experimental results on unbalanced datasets with diverse characteristics have demonstrated that the proposed method could generate a random forest model with higher performance than the random forests generated by Breiman's method.

Citation Download Citation

Baoxun Xu, Yunming Ye, Qiang Wang, Junjie Li, and Xiaojun Chen "Random forest using tree selection method to classify unbalanced data", Proc. SPIE 8334, Fourth International Conference on Digital Image Processing (ICDIP 2012), 83344F (8 June 2012); https://doi.org/10.1117/12.970545

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available