Journal Article

FZJ-2018-02561

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png A simple plug-in bagging ensemble based on threshold-moving for classifying binary and multiclass imbalanced data

Collell, G. ; Prelec, D. ; Patil, K. (Corresponding author)FZJ*

2018
Elsevier Amsterdam

This record in other databases:      

Please use a persistent id in citations: http://hdl.handle.net/2128/18233  doi:10.1016/j.neucom.2017.08.035

Abstract: Class imbalance presents a major hurdle in the application of classification methods. A commonly taken approach is to learn ensembles of classifiers using rebalanced data. Examples include bootstrap averaging (bagging) combined with either undersampling or oversampling of the minority class examples. However, rebalancing methods entail asymmetric changes to the examples of different classes, which in turn can introduce their own biases. Furthermore, these methods often require specifying the performance measure of interest a priori, i.e., before learning. An alternative is to employ the threshold moving technique, which applies a threshold to the continuous output of a model, offering the possibility to adapt to a performance measure a posteriori, i.e., a plug-in method. Surprisingly, little attention has been paid to this combination of a bagging ensemble and threshold-moving. In this paper, we study this combination and demonstrate its competitiveness. Contrary to the other resampling methods, we preserve the natural class distribution of the data resulting in well-calibrated posterior probabilities. Additionally, we extend the proposed method to handle multiclass data. We validated our method on binary and multiclass benchmark data sets by using both, decision trees and neural networks as base classifiers. We perform analyses that provide insights into the proposed method.

---

Contributing Institute(s):

1. Gehirn & Verhalten (INM-7)

Research Program(s):

1. 899 - ohne Topic (POF3-899) (POF3-899)

Appears in the scientific report 2018

---

Database coverage:
; ; ; Current Contents - Engineering, Computing and Technology ; Ebsco Academic Search ; IF < 5 ; JCR ; SCOPUS ; Science Citation Index Expanded ; Thomson Reuters Master Journal List ; Web of Science Core Collection

---