Modified Kohonen network algorithm for selection of the initial centres of Gustafson-Kessel algorithm in credit scoring

Credit risk assessment has become an important topic in financial risk administration. Fuzzy clustering analysis has been applied in credit scoring. Gustafson-Kessel (GK) algorithm has been utilised to cluster creditworthy customers as against non-creditworthy ones. A good clustering analysis implem...

Full description

Saved in:
Bibliographic Details
Main Authors: Sameer, Fadhaa Othman, Abu Bakar, Mohd Rizam
Format: Article
Language:English
Published: Universiti Putra Malaysia Press 2017
Online Access:http://psasir.upm.edu.my/id/eprint/51605/1/06%20JST%20Vol%2025%20%281%29%20Jan%20%202017_0591-2015_pg77-90.pdf
http://psasir.upm.edu.my/id/eprint/51605/
http://www.pertanika.upm.edu.my/Pertanika%20PAPERS/JST%20Vol.%2025%20(1)%20Jan.%202017/06%20JST%20Vol%2025%20(1)%20Jan%20%202017_0591-2015_pg77-90.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Universiti Putra Malaysia
Language: English
Description
Summary:Credit risk assessment has become an important topic in financial risk administration. Fuzzy clustering analysis has been applied in credit scoring. Gustafson-Kessel (GK) algorithm has been utilised to cluster creditworthy customers as against non-creditworthy ones. A good clustering analysis implemented by good Initial Centres of clusters should be selected. To overcome this problem of Gustafson-Kessel (GK) algorithm, we proposed a modified version of Kohonen Network (KN) algorithm to select the initial centres. Utilising similar degree between points to get similarity density, and then by means of maximum density points selecting; the modified Kohonen Network method generate clustering initial centres to get more reasonable clustering results. The comparative was conducted using three credit scoring datasets: Australian, German and Taiwan. Internal and external indexes of validity clustering are computed and the proposed method was found to have the best performance in these three data sets.