Using Neural Network Rule Extraction and Decision Tables for Credit-Risk Evaluation

Abstract

C redit-risk evaluation is a very challenging and important management science problem in the domain of financial analysis. Many classification methods have been suggested in the literature to tackle this problem. Neural networks, especially, have received a lot of attention because of their universal approximation property. However, a major drawback associated with the use of neural networks for decision making is their lack of explanation capability. While they can achieve a high predictive accuracy rate, the reasoning behind how they reach their decisions is not readily available. In this paper, we present the results from analysing three real-life credit-risk data sets using neural network rule extraction techniques. Clarifying the neural network decisions by explanatory rules that capture the learned knowledge embedded in the networks can help the credit-risk manager in explaining why a particular applicant is classified as either bad or good. Furthermore, we also discuss how these rules can be visualized as a decision table in a compact and intuitive graphical format that facilitates easy consultation. It is concluded that neural network rule extraction and decision tables are powerful management tools that allow us to build advanced and userfriendly decision-support systems for credit-risk evaluation.

The Association Between the Disclosure and the Realization of Information Security Risk Factors

“…We use a decision tree for the following two reasons. First, the inherent transparency and interpretability of decision tree models help users follow the path of the tree and understand the classification rules step by step (e.g., Baesens et al 2003;Zhou and Jiang 2004). Second, studies such as Goto et al (2008), Long et al (1993), and Rudolfer et al (1999) have shown that decision tree models have a better accuracy rate for our sampling method than logistic models.…”

APATE: A novel approach for automated credit card transaction fraud detection using network-based extensions

“…Data mining is a well-known and often suitable solution to big data problems involving risk such as credit risk modelling (Baesens et al, 2003), churn 5 A C C E P T E D M A N U S C R I P T…”

Realization of feature description systems for clusters by rule generation based on genetic programming and its applications

“…The German credit data are obtainable from a website. The data consist of 1000 records of personal loans, and the input variables for one record include 7 numerical data and 13 categorical data [3,22].…”

“…Even though there are systems oriented to linguistic expressions, such as ID3, these systems usually need multiple clusters to distinguish between clusters (called pair-samples). For example, if we find a linguistic description for a cluster having "good" as a prescribed value, we also need a cluster having "bad" as a prescribed value [3][4][5][6]. Thus, these methods cannot be directly used to describe the features of a single isolated cluster.…”