A fuzzy decision support system for credit scoring

Joshua Ignatius, Adel Hatami-Marbini, Amirah Rahman, Lalitha Dhamotharan, Pegah Khoshnevis

Research output: Contribution to journalArticlepeer-review

43 Citations (Scopus)

Abstract

Credit score is a creditworthiness index, which enables the lender (bank and credit card companies) to evaluate its own risk exposure toward a particular potential customer. There are several credit scoring methods available in the literature, but one that is widely used is the FICO method. This method provides a score ranging from 300 to 850 as a fast filter for high-volume complex credit decisions. However, it falls short in the aspect of a decision support system where revised scoring can be achieved to reflect the borrower’s strength and weakness in each scoring dimension, as well as the possible trade-offs made to maintain one’s lending risk. Hence, this study discusses and develops a decision support tool for credit score model based on multi-criteria decision-making principles. In the proposed methodology, criteria weights are generated by fuzzy AHP. Fuzzy linguistic theory is applied in AHP to describe the uncertainties and vagueness arising from human subjectivity in decision making. Finally, drawing from the risk distance function, TOPSIS is used to rank the alternatives based on the least risk exposure. A sensitivity analysis is also demonstrated by the proposed fuzzy AHP-TOPSIS method.

Original languageEnglish
Pages (from-to)921-937
Number of pages17
JournalNeural Computing and Applications
Volume29
Issue number10
Early online date26 Sep 2016
DOIs
Publication statusPublished - 1 May 2018
Externally publishedYes

Fingerprint

Dive into the research topics of 'A fuzzy decision support system for credit scoring'. Together they form a unique fingerprint.

Cite this