• International Journal of Quality Innovation
• /
• 제5권2호
• /
• pp.1-16
• /
• 2004

Credit scoring is widely used to make credit decisions, to reduce the cost of credit analysis and enable faster decisions. However, traditional credit scoring models do not account for the influence of noises. This study proposes a robust credit scoring system based on Mahalanobis-Taguchi System (MTS). The MTS, primary proposed by Taguchi, is a diagnostic and forecasting method using multivariate data. The proposed approach's effectiveness is demonstrated by using real case data from a large Taiwanese bank. The results reveal that the robust credit scoring system can be successfully implemented using MTS technique.

• 지능정보연구
• /
• 제9권2호
• /
• pp.135-154
• /
• 2003

Credit scoring system (CSS) starts from an analysis of delinquency trend of each individual or industry. This paper conducts a research on credit card delinquency of bank customers as a preliminary step for building effective credit scoring system to prevent excess loan or bad credit status. To serve this purpose, we use association rules as a rule generating data mining technique. Specifically, we generate sets of rules of customers who are in bad credit status because of delinquency by association rule mining. We expect that the sets of rules generated by association rule mining could act as an estimator of good or bad credit status classifier and basic component of early warning system.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2003년도 Proceeding
• /
• pp.149-154
• /
• 2003

Credit scoring system (CSS) starts from an analysis of delinquency trend of each individual or industry. This paper conducts a research on credit card delinquency of bank customers as a preliminary step for building effective credit scoring system to prevent excess loan or bad credit status. To serve this purpose, we use association rules that ore generating method. Specifically, we generate sets of rules of customers who are in bad credit status because of delinquency by using association rules. We expect that the sets of rules generated by association rules could act as an estimator of good or bad credit status classifier.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2001년도 The Pacific Aisan Confrence On Intelligent Systems 2001
• /
• pp.128-132
• /
• 2001

In this paper, we evaluate and contrast four neural network rule extraction approaches for credit scoring. Experiments are carried our on three real life credit scoring data sets. Both the continuous and the discretised versions of all data sets are analysed The rule extraction algorithms, Neurolonear, Neurorule. Trepan and Nefclass, have different characteristics, with respect to their perception of the neural network and their way of representing the generated rules or knowledge. It is shown that Neurolinear, Neurorule and Trepan are able to extract very concise rule sets or trees with a high predictive accuracy when compared to classical decision tree(rule) induction algorithms like C4.5(rules). Especially Neurorule extracted easy to understand and powerful propositional if -then rules for all discretised data sets. Hence, the Neurorule algorithm may offer a viable alternative for rule generation and knowledge discovery in the domain of credit scoring.

• Journal of the Korean Data and Information Science Society
• /
• 제17권4호
• /
• pp.1085-1090
• /
• 2006

In general, the coarse classifying procedure splits the values of a continuous characteristic into bands and the values of a discrete characteristic into groups of values. Coarse classifying improves the robustness of the credit scoring system but it is complicate and troublesome procedure. Thus, in this paper, we develop a software for coarse classifying by using Visual Basic Language. By using the developed software, we can find the best split easily. Also, this software will help learners to study credit scoring.

• 정보보호학회논문지
• /
• 제29권6호
• /
• pp.1285-1303
• /
• 2019

현재의 신용평가체계에서는 신용정보회사가 개인의 신용정보를 금융기관으로부터 수집하고 이를 기반으로 신용평가를 수행한다. 하지만 이런 신용평가방법은 민감한 신용정보가 하나의 중앙기관에 집중되기에 프라이버시 침해 소지가 있으며, 외부의 공격이 성공할 경우 대규모의 개인정보가 유출될 수 있다. 본 논문에서는 이 문제를 해결하기 위해 개인이 스스로 금융기관으로부터 수집한 신용정보를 바탕으로 신용점수를 계산하고, 이 신용점수가 정상적으로 계산되었음을 영지식 증명과 블록체인으로 증명하는 프라이버시 보장 신용평가방법을 제안한다. 또한 영지식 증명에 이용된 신용정보가 금융기관에서 실제로 제공한 값인지 블록체인을 통해 확인하기 위해, 커밋된 입력에 대해 효율적으로 증명이 가능한 영지식 증명 기법을 제안한다. 이 기법은 Agrawal 등의 기법과 달리 완벽한 영지식성을 제공하며, CRS와 증명의 크기가 작고 증명과 검증 과정이 빠르다. 그리고 제안한 신용평가방법에 실제 환경과 유사한 신용점수 알고리즘을 적용하여 구현 및 실험함으로써 실제 환경에 이용 가능함을 확인하였다.

• 응용통계연구
• /
• 제24권4호
• /
• pp.587-595
• /
• 2011

Credit scoring is an objective and automatic system to assess the credit risk of each customer. The logistic regression model is one of the popular methods of credit scoring to predict the default probability; however, it may not detect possible nonlinear features of predictors despite the advantages of interpretability and low computation cost. In this paper, we propose to use a generalized partially linear model as an alternative to logistic regression. We also introduce modern ensemble technologies such as bagging, boosting and random forests. We compare these methods via a simulation study and illustrate them through a German credit dataset.

• 경영정보학연구
• /
• 제1권1호
• /
• pp.51-61
• /
• 1999

• 경영과학
• /
• 제15권1호
• /
• pp.97-115
• /
• 1998

This study aims to develop the methodology based on the AHP(Analytic Hierarchy Process) of evaluation for commercial paper. commercial paper is the ma product of merchant banks. commercial paper evaluation is annually performed by the credit-evaluation agency. Credit evaluation is performed by the informal judgemental system, which has potential to induce serious inconsistencies in decision-making. We present an objective scoring model which does not suffer from the weakness of the subjective judgement system. The model used is illustrated by analyzing the commercial paper evaluation for the 3 motor-companies(H, K and S motors).

• 국제학술발표논문집
• /
• The 4th International Conference on Construction Engineering and Project Management Organized by the University of New South Wales
• /
• pp.399-406
• /
• 2011

Various construction bonds and warranties critically burden the general contractor. Also, sporadic or cumulative delays of progress payment by the owner can further trap the contractor in a financial quagmire. Facing the possibility of cash flow deficiency and callous response from the banks, most construction firms may become financially incapable of market competition, and attractive project tenders become a bidding game among few deep-pocket players. The downside of such market environment is that the depth of pocket, rather than that of professional competency dictates the choice of market winners. In Taiwan, this has been a potential crisis to the construction industry after the financial crisis which started out since 2008. To encounter this problem, this research will examine the means to better manage the construction industry. Essentially, a credit guarantee system (CGS) is the prime solution to strengthen a bank's confidence in any particular construction firm. Thus establishing a national platform which evaluates and rewards a construction firm's overall credibility is pivotal, and this third-party rated credit can help a bank to render a loan more wisely. Finally, this paper will propose the ideal operating schemes of construction-specific CGS in Taiwan and a credit scoring prototype model for construction industry, as reference for the government and banks, respectively.