| 基于自适应遗传模糊神经网络的信用评估建模 |
| |
| 引用本文： | 熊志斌.基于自适应遗传模糊神经网络的信用评估建模[J].系统仿真学报,2011,23(3):490-496. |
| |
| 作者姓名： | 熊志斌 |
| |
| 作者单位： | 华南师范大学数学科学学院,广州,510631 |
| |
| 基金项目： | 国家自然科学基金，高校博士点科研基金项目，广东省哲社科"十一五"规划项目 |
| |
| 摘 要： | 提出了一种自适应遗传模糊神经网络评估信用风险的模型,该模型在多子群遗传算法基础上,采用带控制参数的动态概率选择与最优保存策略相结合的混合选择策略,根据种群适应度标准差大小动态调整交叉和变异概率,并将BP算子嵌入遗传算法中,构建了多子群自适应遗传BP算法,并利用该算法优化网络的连接权值和模糊参数。将所建模型应用到信用评估中,并与BP神经网络、ANFIS以及遗传神经网络模型预测效果进行比较,结果表明该模型对信用评估具有更好的泛化能力和更高的预测准确度。
|
| 关 键 词： | 遗传算法 混合选择策略 自适应 模糊神经网络 信用评估 |
Credit Evaluation Modelling Based on Self-adaptive Genetic Fuzzy Neural Network |
| |
| XIONG Zhi-bin.Credit Evaluation Modelling Based on Self-adaptive Genetic Fuzzy Neural Network[J].Journal of System Simulation,2011,23(3):490-496. |
| |
| Authors: | XIONG Zhi-bin |
| |
| Institution: | XIONG Zhi-bin(School of Mathematical Sciences,South China Normal University,Guangzhou 510631,China) |
| |
| Abstract: | A self-adaptive genetic fuzzy neural network(SAGFNN) model for evaluating credit risk was proposed.A hybrid selection strategy was employed which combined dynamic selection probability with control parameter with elitist strategy in this model.Crossover probability and mutation probability were self-adjusted according to the standard deviation of population fitness in this model,BP operator was embedded in genetic algorithm at the same time.Multi-population self-adaptive genetic BP algorithm(MSAGBPA) was de... |
| |
| Keywords: | genetic algorithm hybrid selection strategy self-adaptive fuzzy neural network credit evaluation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
