Better classifiers for credit scoring: a comparison study between self organizing maps (SOM) and support vector machine (SVM)

Authors:
Afshin Shahlaii Moghadam;Ali Shalbafzadeh;Mohammad Saraee
Affiliations:
Department of Industrial Engineering, Isfahan University of Technology, Isfahan University of Technology, Isfahan and Tadbir Strategy Builder © Ltd, Tehran, Iran;Department of Industrial Engineering, Isfahan University of Technology, Isfahan University of Technology, Isfahan and Tadbir Strategy Builder © Ltd, Tehran, Iran;Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan University of Technology, Isfahan and School of Computing, Science and Engineering, University of Salfo ...
Venue:
CIT'09 Proceedings of the 3rd International Conference on Communications and information technology
Year:
2009

Citing 4
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Self-organizing maps

Self-organizing maps
A Tutorial on Support Vector Machines for Pattern Recognition

Data Mining and Knowledge Discovery
A novel support vector machine metamodel for business risk identification

PRICAI'06 Proceedings of the 9th Pacific Rim international conference on Artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Credit scoring has become an increasingly important area for financial institutions. Self Organizing Maps and Support Vector Machine are two techniques of data mining which are used in different applications of businesses. In this paper, we use descriptive variables in literatures and criteria which effect on credit of customers in Iran financial institutions. We will evaluate these variables with Multi Criteria Decision Making (MCDM) and take into account the psychological and sociology viewpoints of experts. Next We apply and compare SVM method against SOM method on the credit database. For comparing these two methods we use coincidence matrix and the Type I and Type II errors. We show that they are competitive and most significant in determining the risk of default on bank customers.