摘要
个人信用作为社会信用体系建设的重要部分,将其结合现代计算机理论技术来构建个人信用评分模型一直是研究的热点。本文利用前人遗传算法筛选出来的个人信用相关重要属性,并从这些重要属性的3种分类中依类定性地取出部分属性,结合自适应神经模糊推理系统理论(ANFIS),建立基于遗传算法和ANFIS的个人信用评分模型。对选取的数据实证分析,并与GA-SVM方法的结果作了比较,试验结果表明该模型只需少量重要属性变量就能够有较好的分类效果。
The integration of personal credit, which is an important part of social credit system, with modern computer science and technology in the process of personal credit scoring model construction has become a hot topic. This paper draws upon the important attribute of personal credit which is selected by previous researcher through genetic algorithm and set out to build a personal credit scoring model that is based on the combination of Genetic Algorithm and ANTFIS. The experimental analysis of data and comparison of the results with GA-SVM show that it has better result than GA-SVM with minimal changes of attribute variables.
出处
《福建师大福清分校学报》
2013年第5期1-6,共6页
Journal of Fuqing Branch of Fujian Normal University
基金
福建省教育厅A类科技项目(JA12353)
福建省教育厅A类科技项目(JA12351)
福建师范大学福清分校科研项目(KY2012025)
福建省省属高校科研专项课题(JK2013062)
福建省大学生创新创业训练计划项目(sjcxcy-20421322)
关键词
自适应模糊神经推理系统
遗传算法
信用评分
adaptive neuro fuzzy inference system
genetic algorithm
credit scoring
