資料紹介
資料の原本内容
An Application of the Rough Set Approach to Credit Rating
SungSik Cho(, JaeKyeong Kim((
Abstract
The credit rating represents an assessment of the relative level of risk associated with the timely payments required by the debt obligation. In this paper, we present a new approach to credit rating of customers based on the rough set theory. The concept of a rough set appeared to be an effective tool for the analysis of customer information systems representing knowledge gained by experience. The customer information system describes a set of customers by a set of multi-valued attributes, called condition attributes. The customers are classified into groups of risk subject to an expert’s opinion, called decision attribute. A natural problem of knowledge analysis consists then in discovering relationships, in terms of decision rules, between description of customers by condition attributes and particular decisions. The rough set approach enables one to discover minimal subsets of condition attributes ensuring an acceptable quality of classification of the customers analyzed and to derive decision rules from the customer information system which can be used to support decisions about rating new customers.
Using the rough set approach one analyses only facts hidden in data, it does not need any additional information about data and does not correct inconsistencies manifested in data; instead, rules produced are categorized into certain and possible. A real problem of the evaluation of credit rating by a department store is studied using the rough set approach.
Key words: Credit Rating, Rough Set
Introduction
Credit rating of a credit card customers has been, for a long time, a major preoccupation of university researchers and practitioners. The first approach to rating the credit of credit card customers started with the use of demographic features gathered at new customer entrance. These features were long considered as objective indicators of bad customers.
Recently, new methods of credit rating have been developed using the usage information along with demographic features. There were developed statistical tools based on multivariate statistical methods (e.g. discriminant analysis, cluster analysis) which classify credit customers into groups of bad or good, or calculate a score representing the degree of arrearage risk using those demographic features which considered significant. The most common methods are those of ‘credit scoring’, which establish a discriminant function using some of a customers demographic and usage features, and classify them into high risk or low risk groups (Capon, 1982, Kim et. al. 1999). Later, tools were developed which were based on multi-criteria decision aid methodology. They also classify customers into groups of risk, but circumvent many of the problems that exist when using discriminant analysis. Finally, tools based on artificial intelligence were developed.
This paper presents a new method called the rough set approach for the analysis and evaluation of the credit rating. The concept of a rough set introduced by Pawlak (1982) proved to be an effective tool for the analysis of information systems (customer information system) describing a set of objects (customers) by a set of multi-valued attributes. In particular, in the case where the set of objects is classified subject to an expert’s opinion, this approach enables one to deal with two basic problems of information systems:
How to reduce the set of attributes to a subset ensuring as good approximation of the expert’s classification as the whole set of attributes.
How to derive decision rules from the information systems in view of explaining a decision policy of the expert.
The decision rules derived with the rough set approach are expressed in terms of significant attributes without any redundancy typical for original data. They are based on facts because each decision rule is supported by a set of real examples. In comparison with other...
