ASE-Bagging : An Embedded approach for imbalanced customer credit risk assessment

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalNot applicablepeer-review

2 Scopus Citations
View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Original languageEnglish
Pages (from-to)787-791
Journal / PublicationICIC Express Letters, Part B: Applications
Volume2
Issue number4
Publication statusPublished - Aug 2011

Abstract

For the problem of customer credit risk assessment, the training data are often imbalanced in class distribution practically, which significantly influences the performance of assessment. Nevertheless, in this study, we propose an embedded approach, Advanced Sampling Embedded Bagging (ASE-Bagging), for the imbalanced customer credit risk assessment. ASE-Bagging integrates Synthetic Minority Over-sampling Technique (SMOTE) with bagging in order to solve the imbalanced data problem. Two real-world credit data sets are used for evaluating the proposed method. And the empirical results reveal that the proposed ASE-Bagging is a very promising approach for the imbalanced customer credit risk assessment. © 2011 ISSN 2185-2766.

Research Area(s)

  • Advanced sampling, Bagging, Customer credit risk assessment, Ensemble learning, SMOTE

Citation Format(s)