DATA MINING VIA MULTIPLE CRITERIA LINEAR PROGRAMMING: APPLICATIONS IN CREDIT CARD PORTFOLIO MANAGEMENT

2002 ◽  
Vol 01 (01) ◽  
pp. 131-151 ◽  
Author(s):  
YONG SHI ◽  
YI PENG ◽  
WEIXUAN XU ◽  
XIAOWO TANG
Keyword(s):  
Data Mining ◽  
Linear Programming ◽  
Portfolio Management ◽  
Credit Card ◽  
Real Life ◽  
Multiple Criteria ◽  
Business Decisions ◽  
Business World ◽  
Linear Discriminant ◽  
Technology Tool

Data mining becomes a cutting-edge information technology tool in today's competitive business world. It helps the company discover previously unknown, valid, and actionable information from various and large databases for crucial business decisions. This paper provides a promising approach of data mining to classify the credit cardholders' behavior through multiple criteria linear programming. After reviewing the history of linear discriminant analyses, we will describe first a model for classifying two-group (e.g. bad or good) credit cardholder behaviors, and then a three-group (e.g. bad, normal, or good) credit model. Besides the discussion of the modeling structure, we will utilize the well-known commercial software package SAS to implement this technology by using a real-life credit card data warehouse. A number of potential business and financial applications will be finally summarized.

Introduction to Data Mining Techniques via Multiple Criteria Optimization Approaches and Applications

2008 ◽  
pp. 26-49 ◽  
Author(s):  
Yong Shi ◽  
Yi Peng ◽  
Gang Kou ◽  
Zhengxin Chen
Keyword(s):  
Data Mining ◽  
Linear Programming ◽  
Credit Card ◽  
Programming Model ◽  
Real Life ◽  
Multiple Criteria ◽  
Linear Programming Model ◽  
Multiple Criteria Optimization ◽  
Research Challenges ◽  
Challenges And Opportunities

This chapter provides an overview of a series of multiple criteria optimization-based data mining methods, which utilize multiple criteria programming (MCP) to solve data mining problems, and outlines some research challenges and opportunities for the data mining community. To achieve these goals, this chapter first introduces the basic notions and mathematical formulations for multiple criteria optimization-based classification models, including the multiple criteria linear programming model, multiple criteria quadratic programming model, and multiple criteria fuzzy linear programming model. Then it presents the real-life applications of these models in credit card scoring management, HIV-1 associated dementia (HAD) neuronal dam-age and dropout, and network intrusion detection. Finally, the chapter discusses research challenges and opportunities.

CLASSIFYING CREDIT CARD ACCOUNTS FOR BUSINESS INTELLIGENCE AND DECISION MAKING: A MULTIPLE-CRITERIA QUADRATIC PROGRAMMING APPROACH

2005 ◽  
Vol 04 (04) ◽  
pp. 581-599 ◽  
Author(s):  
YONG SHI ◽  
YI PENG ◽  
GANG KOU ◽  
ZHENGXIN CHEN
Keyword(s):  
Decision Making ◽  
Quadratic Programming ◽  
Portfolio Management ◽  
Business Intelligence ◽  
Credit Card ◽  
Real Life ◽  
Multiple Criteria ◽  
Programming Approach ◽  
Support Vector ◽  
Linear Discriminant

A major challenge in credit card portfolio management is to classify and predict credit cardholders' behaviors in a reliable precision because cardholders' behaviors are rather dynamic in nature. This is crucial for creditors because it allows them to take proactive actions and minimize charge-off and bankruptcy losses. Although the methods used in the area of credit portfolio management have improved significantly, the demand for alternative and sophisticated analytical tools is still strong. The objective of this paper is to propose a multiple criteria quadratic programming (MCQP) to classify credit card accounts for business intelligence and decision making. MCQP is intended to predict credit cardholders' behaviors from a nonlinear perspective that is justifiable because both the objective functions and constraints in credit card accounts classification may be nonlinear. Using a real-life credit card dataset from a major US bank, the MCQP method is compared with popular and similar classification methods: linear discriminant analysis, decision tree, multiple criteria linear programming, support vector machine, and neural network. The results indicate that MCQP is a promising business intelligence method in credit card portfolio management.

Introduction to Data Mining Techniques via Multiple Criteria Optimization Approaches and Applications

2007 ◽  
pp. 242-275 ◽  
Author(s):  
Yong Shi ◽  
Yi Peng ◽  
Gang Kou ◽  
Zhengxin Chen
Keyword(s):  
Data Mining ◽  
Linear Programming ◽  
Credit Card ◽  
Programming Model ◽  
Real Life ◽  
Multiple Criteria ◽  
Linear Programming Model ◽  
Multiple Criteria Optimization ◽  
Research Challenges ◽  
Challenges And Opportunities

This chapter provides an overview of a series of multiple criteria optimization-based data mining methods, which utilize multiple criteria programming (MCP) to solve data mining problems, and outlines some research challenges and opportunities for the data mining community. To achieve these goals, this chapter first introduces the basic notions and mathematical formulations for multiple criteria optimization-based classification models, including the multiple criteria linear programming model, multiple criteria quadratic programming model, and multiple criteria fuzzy linear programming model. Then it presents the real-life applications of these models in credit card scoring management, HIV-1 associated dementia (HAD) neuronal dam-age and dropout, and network intrusion detection. Finally, the chapter discusses research challenges and opportunities.

Decision Rule Extraction for Regularized Multiple Criteria Linear Programming Model

2011 ◽  
Vol 7 (3) ◽  
pp. 88-101 ◽  
Author(s):  
DongHong Sun ◽  
Li Liu ◽  
Peng Zhang ◽  
Xingquan Zhu ◽  
Yong Shi
Keyword(s):  
Linear Programming ◽  
Portfolio Management ◽  
Real World ◽  
Credit Card ◽  
Extraction Method ◽  
Large Scale ◽  
Programming Model ◽  
Rule Extraction ◽  
Multiple Criteria ◽  
Data Sets

Due to the flexibility of multi-criteria optimization, Regularized Multiple Criteria Linear Programming (RMCLP) has received attention in decision support systems. Numerous theoretical and empirical studies have demonstrated that RMCLP is effective and efficient in classifying large scale data sets. However, a possible limitation of RMCLP is poor interpretability and low comprehensibility for end users and experts. This deficiency has limited RMCLP’s use in many real-world applications where both accuracy and transparency of decision making are required, such as in Customer Relationship Management (CRM) and Credit Card Portfolio Management. In this paper, the authors present a clustering based rule extraction method to extract explainable and understandable rules from the RMCLP model. Experiments on both synthetic and real world data sets demonstrate that this rule extraction method can effectively extract explicit decision rules from RMCLP with only a small compromise in performance.

Introduction to Data Mining Techniques via Multiple Criteria Optimization Approaches and Applications

2009 ◽  
pp. 1-25
Author(s):  
Yong Shi ◽  
Yi Peng ◽  
Gang Kou ◽  
Zhengxin Chen
Keyword(s):  
Data Mining ◽  
Linear Programming ◽  
Programming Model ◽  
Neuronal Damage ◽  
Real Life ◽  
Multiple Criteria ◽  
Linear Programming Model ◽  
Multiple Criteria Optimization ◽  
Research Challenges ◽  
Challenges And Opportunities

This chapter provides an overview of a series of multiple criteria optimization-based data mining methods, which utilize multiple criteria programming (MCP) to solve data mining problems, and outlines some research challenges and opportunities for the data mining community. To achieve these goals, this chapter first introduces the basic notions and mathematical formulations for multiple criteria optimization- based classification models, including the multiple criteria linear programming model, multiple criteria quadratic programming model, and multiple criteria fuzzy linear programming model. Then it presents the real-life applications of these models in credit card scoring management, HIV-1 associated dementia (HAD) neuronal damage and dropout, and network intrusion detection. Finally, the chapter discusses research challenges and opportunities.

CLASSIFICATIONS OF CREDIT CARDHOLDER BEHAVIOR BY USING FUZZY LINEAR PROGRAMMING

2004 ◽  
Vol 03 (04) ◽  
pp. 633-650 ◽  
Author(s):  
JING HE ◽  
XIANTAO LIU ◽  
YONG SHI ◽  
WEIXUAN XU ◽  
NIAN YAN
Keyword(s):  
Linear Programming ◽  
Portfolio Management ◽  
Credit Card ◽  
Real Life ◽  
Research Direction ◽  
Fuzzy Linear Programming ◽  
Research Topics ◽  
Score System ◽  
Main Research ◽  
Heuristic Classification

Behavior analysis of credit cardholders is one of the main research topics in credit card portfolio management. Usually, the cardholder's behavior, especially bankruptcy, is measured by a score of aggregate attributes that describe cardholder's spending history. In real-life practice, statistics and neural networks are the major players to calculate such a score system for prediction. Recently, various multiple linear programming-based classification methods have been promoted for analyzing credit cardholders' behaviors. As a continuation of this research direction, this paper proposes a heuristic classification method by using the fuzzy linear programming (FLP) to discover the bankruptcy patterns of credit cardholders. Instead of identifying a compromise solution for the separation of credit cardholder behaviors, this approach classifies the credit cardholder behaviors by seeking a fuzzy (satisfying) solution obtained from a fuzzy linear program. In this paper, a real-life credit database from a major US bank is used for empirical study which is compared with the results of known multiple linear programming approaches.

Decision Rule Extraction for Regularized Multiple Criteria Linear Programming Model

2013 ◽  
pp. 245-258
Author(s):  
DongHong Sun ◽  
Li Liu ◽  
Peng Zhang ◽  
Xingquan Zhu ◽  
Yong Shi
Keyword(s):  
Linear Programming ◽  
Portfolio Management ◽  
Real World ◽  
Credit Card ◽  
Extraction Method ◽  
Large Scale ◽  
Rule Extraction ◽  
Multiple Criteria ◽  
Customer Relationship ◽  
Data Sets

Due to the flexibility of multi-criteria optimization, Regularized Multiple Criteria Linear Programming (RMCLP) has received attention in decision support systems. Numerous theoretical and empirical studies have demonstrated that RMCLP is effective and efficient in classifying large scale data sets. However, a possible limitation of RMCLP is poor interpretability and low comprehensibility for end users and experts. This deficiency has limited RMCLP’s use in many real-world applications where both accuracy and transparency of decision making are required, such as in Customer Relationship Management (CRM) and Credit Card Portfolio Management. In this paper, the authors present a clustering based rule extraction method to extract explainable and understandable rules from the RMCLP model. Experiments on both synthetic and real world data sets demonstrate that this rule extraction method can effectively extract explicit decision rules from RMCLP with only a small compromise in performance.

scholarly journals Association rule hiding using integer linear programming

2021 ◽  
Vol 11 (4) ◽  
pp. 3451
Author(s):  
Suma B. ◽  
Shobha G.
Keyword(s):  
Data Mining ◽  
Linear Programming ◽  
Integer Linear Programming ◽  
Association Rule ◽  
Real Life ◽  
Linear Program ◽  
Integer Linear Program ◽  
Optimization Approach ◽  
Integer Linear Program Formulation ◽  
The Impact

<span>Privacy preserving data mining has become the focus of attention of government statistical agencies and database security research community who are concerned with preventing privacy disclosure during data mining. Repositories of large datasets include sensitive rules that need to be concealed from unauthorized access. Hence, association rule hiding emerged as one of the powerful techniques for hiding sensitive knowledge that exists in data before it is published. In this paper, we present a constraint-based optimization approach for hiding a set of sensitive association rules, using a well-structured integer linear program formulation. The proposed approach reduces the database sanitization problem to an instance of the integer linear programming problem. The solution of the integer linear program determines the transactions that need to be sanitized in order to conceal the sensitive rules while minimizing the impact of sanitization on the non-sensitive rules. We also present a heuristic sanitization algorithm that performs hiding by reducing the support or the confidence of the sensitive rules. The results of the experimental evaluation of the proposed approach on real-life datasets indicate the promising performance of the approach in terms of side effects on the original database.</span>

scholarly journals Predicting Auditor Changes Using Financial Distress Variables And The Multiple Criteria Linear Programming (MCLP) And Other Data Mining Approaches

2011 ◽  
Vol 27 (5) ◽  
pp. 73 ◽  
Author(s):  
Wikil Kwak ◽  
Susan Eldridge ◽  
Yong Shi ◽  
Gang Kou
Keyword(s):  
Data Mining ◽  
Linear Programming ◽  
Financial Distress ◽  
Multiple Criteria ◽  
Financial Statement ◽  
Font Size ◽  
Explanatory Variables ◽  
Data Mining Approach ◽  
Starting Point ◽  
Accuracy Rates

<span style="font-family: Times New Roman; font-size: small;"> </span><h1 style="margin: 0in 0.5in 0pt; text-align: justify; page-break-after: auto; mso-pagination: none;"><span style="font-family: Times New Roman;"><span style="color: black; font-size: 10pt; mso-themecolor: text1;">Our study evaluates a multiple criteria linear programming (MCLP) </span><span style="color: black; font-size: 10pt; mso-themecolor: text1; mso-fareast-language: KO;">and other </span><span style="color: black; font-size: 10pt; mso-themecolor: text1;">data mining approach</span><span style="color: black; font-size: 10pt; mso-themecolor: text1; mso-fareast-language: KO;">es</span><span style="color: black; font-size: 10pt; mso-themecolor: text1;"> </span><span style="color: black; font-size: 10pt; mso-themecolor: text1; mso-fareast-language: KO;">to predict auditor changes using a portfolio of financial statement measures to capture financial distress</span><span style="color: black; font-size: 10pt; mso-themecolor: text1;">.<span style="mso-spacerun: yes;"> </span>The results of the MCLP approach and the other data mining approaches show that these methods perform</span><span style="color: black; font-size: 10pt; mso-themecolor: text1; mso-fareast-language: KO;"> reasonably well to predict auditor changes </span><span style="color: black; font-size: 10pt; mso-themecolor: text1;">using financial distress variables.</span><span style="color: black; font-size: 10pt; mso-themecolor: text1; mso-fareast-language: KO;"><span style="mso-spacerun: yes;"> </span>Overall accuracy rates are more than 60 percent, and true positive rates exceed 80 percent.<span style="mso-spacerun: yes;"> </span>Our study is designed to establish a starting point for auditor-change prediction using financial distress variables.<span style="mso-spacerun: yes;"> </span>Further research should incorporate additional explanatory variables and a longer study period to improve prediction rates.</span></span></h1><span style="font-family: Times New Roman; font-size: small;"> </span>

Sign in / Sign up

Export Citation Format

Share Document