A Multi-criteria Convex Quadratic Programming model for credit data analysis

• Published in: Decision Support Systems Vol. 44; no. 4; pp. 1016 - 1030
• Main Authors: Peng, Yi, Kou, Gang, Shi, Yong, Chen, Zhengxin
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.03.2008; Elsevier Science; Elsevier Sequoia S.A
• Subjects: Applied sciences; Classification; Computer science; control theory; systems; Credit; Data mining; Data processing. List processing. Character string processing; Decision theory. Utility theory; Exact sciences and technology; Knowledge; Mathematical programming; Memory organisation. Data processing; Multi-criteria Convex Quadric Programming (MCQP); Multiple criteria decision making; Operational research and scientific management; Operational research. Management science; Portfolio theory; Quadratic programming; Software; Studies; Multiple criteria decision making; Multi-criteria Convex Quadric Programming (MCQP); Data mining; Classification; Mathematical programming; Quadric; Scalability; Multicriteria analysis; Information extraction; Modeling; Convex programming; Credit; Vector support machine; Knowledge discovery; Data analysis; Decision making; Banking; Data processing; Quadratic programming; Computational complexity; Global solution; Kernel function; Non linear model; Optimal solution; Data models; Time complexity; Convex analysis
• ISSN: 0167-9236; 1873-5797
• DOI: 10.1016/j.dss.2007.12.001

Speed and scalability are two essential issues in data mining and knowledge discovery. This paper proposed a mathematical programming model that addresses these two issues and applied the model to Credit Classification Problems. The proposed Multi-criteria Convex Quadric Programming (MCQP) model is highly efficient (computing time complexity O( n 1.5–2)) and scalable to massive problems (size of O(10 9)) because it only needs to solve linear equations to find the global optimal solution. Kernel functions were introduced to the model to solve nonlinear problems. In addition, the theoretical relationship between the proposed MCQP model and SVM was discussed.