A Model for Recognizing Key Factors and Applications Thereof to Engineering

This paper presents an approach to recognize key factors in data classification. Using collinearity diagnostics to delete the factors of repeated information and Logistic regression significant discriminant to select the factors which can effectively distinguish the two kinds of samples, this paper creates a model for recognizing key factors. The proposed model is demonstrated by using the 2044 observations in finical engineering. The experimental results demonstrate that the 13 indicators such as “marital status,” “net income of borrower,” and “Engel's coefficient” are the key factors to distinguish the good customers from the bad customers. By analyzing the experimental results, the performance of the proposed model is verified. Moreover, the proposed method is simple and easy to be implemented.

[1]  Tien-Chin Wang,et al.  Applying Rough Sets Theory to Corporate Credit Ratings , 2006, 2006 IEEE International Conference on Service Operations and Logistics, and Informatics.

[2]  Zhenyuan Wang,et al.  Using Non-Additive Measure for Optimization-Based Nonlinear Classification , 2012 .

[3]  Jian-Liung Chen,et al.  Combining Biometric Fractal Pattern and Particle Swarm Optimization-Based Classifier for Fingerprint Recognition , 2010 .

[4]  Bhekisipho Twala,et al.  Multiple classifier application to credit risk assessment , 2010, Expert Syst. Appl..

[5]  Chi Guotai,et al.  A Credit Risk Evaluation Index Screening Model of Petty Loans for Small Private Business and Its Application -- The Case of Chinese Small Private Business Data , 2013 .

[6]  Steven Finlay,et al.  Multiple classifier architectures and their application to credit risk assessment , 2011, Eur. J. Oper. Res..

[7]  Ping Guo,et al.  Adaptive Multilevel Kernel Machine for Scene Classification , 2013 .

[8]  Young-Chan Lee,et al.  A practical approach to credit scoring , 2008, Expert Syst. Appl..

[9]  Mats Gyllenberg,et al.  Bayesian unsupervised classification framework based on stochastic partitions of data and a parallel search strategy , 2009, Adv. Data Anal. Classif..

[10]  Soner Akkoç,et al.  An empirical comparison of conventional techniques, neural networks and the three stage hybrid Adaptive Neuro Fuzzy Inference System (ANFIS) model for credit scoring analysis: The case of Turkish credit card data , 2012, Eur. J. Oper. Res..

[11]  C. K. Chu,et al.  Predicting issuer credit ratings using a semiparametric method , 2010 .

[12]  Sorin Moga,et al.  Combined pattern search optimization of feature extraction and classification parameters in facial recognition , 2011, Pattern Recognit. Lett..

[13]  A. C. Rencher.  Linear models in statistics , 1999 .

[14]  Kenneth Carling,et al.  Corporate credit risk modeling and the macroeconomy , 2007 .