Classifiers consensus system approach for credit scoring
- Submitting institution
-
Brunel University London
- Unit of assessment
- 12 - Engineering
- Output identifier
- 110-122354-9425
- Type
- D - Journal article
- DOI
-
10.1016/j.knosys.2016.04.013
- Title of journal
- Knowledge Based Systems
- Article number
- -
- First page
- 89
- Volume
- 104
- Issue
- -
- ISSN
- 0950-7051
- Open access status
- Compliant
- Month of publication
- April
- Year of publication
- 2016
- URL
-
https://bura.brunel.ac.uk/handle/2438/12571
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
1
- Research group(s)
-
4 - Sensors & Digital Systems
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper presents an improved classifier system using classifier consensus method to combine multiple classification algorithms. The combined classifiers are neural networks, support vector machines, random forests, decision trees and naïve Bayes. The classifier consortium is tested on credit scoring data, the classifier performance is compared to Linear Regression and multivariate adaptive regression splines. Experimental results demonstrate the proposed method improves prediction accuracy against all other classifiers. The model was validated over five real-world credit scoring data sets. The method is being tested on real credit scoring data from a bank in the Middle East (contact details available on request).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -