基于案例推理循环优化的个人信用评分方法分析-国际贸易学专业论文.docxVIP

Ab Abstract - - II - Abstract With the accelerating world economic integration process, risk control is more important to commercial bank. As one of the essential parts of bank assets business, the level of personal credit is related to the general requirements of economy capital.Therefore, effective recognit ion and measurement for personal risk will become the key factor of reasonable risk control for the commercial bank. At present, most credit scoring models globally are built on statistics or artificial intelligence, which are all double-edged. Statistical models offer tests but fall short of accuracy, while artificial intelligence achieves high accuracy but fall short of explanatory function. Furthermore, these models are challenged by biased sample and dynamic changing of credit samples which need to be solved urgently. Case-based reasoning (CBR) simulates the cognit ive process of humanity, enjoying a strong theoretical background and varied applications and has potential to solve these problems.A tradit ional CBR credit scoring system is built to analyze adavantages and disadvantages of the CBR when it is used in credit scoring. The disadvantages are mainly in the progress of CBR cycle. To solve these problems, a normalization algorithm for mult iple attribute index, a hybrid mult iple distance measurement algorithm and a calculation method with variable feature are introduced in the case retrieval to optimize it and improve robustness of model. What`s more, Bayesian algorithm is employed as the reusing method to use the valid data fully. The results of an experiment shows that the improved CBR cycle system not only has a better performance on solving problems of biased sample and dynamic changing of credit samples; the optimized CBR cycle system is more accurate and has obvious progress in stability and explanatory function comparing to traditional system, but also has a better control on misclassification cost. Keywords: credit scoring, Biased Samp