Data Mining in Credit Card Portfolio Management: A Multiple Criteria Decision Making Approach

Shi, Yong; Wise, Morgan; Luo, Ming; Lin, Yuxia

doi:10.1007/978-3-642-56680-6_39

Cited by 87 publications

(51 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regardless of the type of drilling rigs used in Iran and according to the data obtained from previous drilling operations, the objective of this paper is to provide a model for predicting the penetration rate. By comparing and studying the useful factors about the mud, the type of drilling, hydraulics, the rig power and, the study of the conditions can be achieved using modern modeling techniques [34][35][36][37][38][39][40][41][42][43][44][45].…”

Section: Support Vector Machine (Svm)mentioning

confidence: 99%

A Model to Predict the Performance of Drilling Operation Using SVM Approach in One of the Iranian Oilfields

2018

International Journal of Petroleum and Petrochemical Engineerin

View full text Add to dashboard Cite

Section: Support Vector Machine (Svm)mentioning

confidence: 99%

A Model to Predict the Performance of Drilling Operation Using SVM Approach in One of the Iranian Oilfields

2018

International Journal of Petroleum and Petrochemical Engineerin

View full text Add to dashboard Cite

“…MCLP classification model is first presented by Shi et al [1]. A two-class MCLP model can be depicted as follows:…”

Section: Two-class Mclp Classification Modelmentioning

confidence: 99%

“…In data mining field, Multiple-Criteria Linear Programming (MCLP) classification method is an outstanding classification tool [1][2][3][4][5]. But, just like many other classification tools, its computation efficiency is sometimes low when faced with large and high-dimension datasets.…”

Section: Introductionmentioning

confidence: 99%

A Dynamic Committee Scheme on Multiple-Criteria Linear Programming Classification Method

Zhu

Shi²,

Li³

et al. 2007

Computational Science – ICCS 2007

Self Cite

View full text Add to dashboard Cite

Abstract. This paper aims to provide a scheme for effectively and efficiently finding an approximately optimal example size with respect to a given dataset when using Multiple-Criteria Linear Programming (MCLP) classification method. By integrating techniques of both progressive sampling and classification committee, it designs a dynamic classification committee scheme for MCLP. The experimental results have shown that our idea is feasible and the scheme is effective and efficient for exploring an approximately optimal sample size. The empirical results also help us to further investigate some general specialties of MCLP, such as the more general function expressions reflecting the relationship between accuracy and sample size, and between computing cost and sample size.

show abstract

“…If one hasa scoringsystem thatgivesa score to each member of the population, then one can use these measures to describe how differentt he scoresof the good and the bad are. Thus these approaches can be used only for credit-scoring systems that actuallygive a s core, like t he r egression approaches or linear programming [7,8,9]. Theycannot be used for the credit-scoring systems that group, like classification trees, orw here a score is not explicit, like neural networks.…”

Section: Separation Measures: Mahalanobis Distance and Kolmogorov-smmentioning

confidence: 99%

Measuring Scorecard Performance

Yang

Wang

Bai

et al. 2004

Computational Science - ICCS 2004

View full text Add to dashboard Cite

Abstract. Inthispaper,we look at ways t om easure t he classification performance of a scoring system and the overall characteristics of a scorecard. Westickto the idea that we will measure the scoring system by howw ell it classifies, w hich are s till problems in measuring its performance. This is because there are different ways to define the misclassification rate mainly due to the sample that we use to check this rate. If we test how good the system is on the sample of customers we used to build the system, the results will be better than that we did the test on another sample. This idea is illustrated in this paper. Two measures, Mahalanobis distance and KS score, areused in the paper.Keywords: credit-scoring s ystems,measuring s corecard,c lassification, holdout, mahalanobisdistance, KS score. 1IntroductionHaving built a credit or behavioral scorecard, the obvious question is, "How good is it ?" this begs the question of whatw em eanbygood. The obvious answer isin distinguishing the good from the bad becausewewant to treat these groupsin different ways in credit-scoring systems---for example,accepting the former for credit and rejecting the latter. Behavioral scoringsystems are used in a more subtle way,but even if we stick to the idea that wewillm easure the scoring system byhow well it classifies, there are still problems in measuring its performance. This is because there are different ways tod efinethe misclassificationrate, mainlydue to the sample that we use to check this rate. Ifwe test how good the system is on the sample of customers weused to build the system, the results will bem ust better than if we did the test on another sample. This must followbecause built intothe classification system are someo fthe nuances of that data set that do not appear in other data sets. Thus section 2 looks at how to test the classification rate using a sample, called the holdout sample, separate from the one used to build the scoring system. This is a very common thing to do in the credit-scoringi ndustrybecause of the availability of very large samples of past customers, but it is wasteful of data in that one does not use all the information available to help build the best scoring system. There are times, however, whenthe amount of data is limited, for example, when one is building a system for a completelynewgroup of customers or products. In the case, one can test

show abstract

Data Mining in Credit Card Portfolio Management: A Multiple Criteria Decision Making Approach

Cited by 87 publications

References 4 publications

A Model to Predict the Performance of Drilling Operation Using SVM Approach in One of the Iranian Oilfields

A Model to Predict the Performance of Drilling Operation Using SVM Approach in One of the Iranian Oilfields

A Dynamic Committee Scheme on Multiple-Criteria Linear Programming Classification Method

Measuring Scorecard Performance

Contact Info

Product

Resources

About