基于大数据分析的银行不良信贷风险模型

Risk Model of Bank Bad Credit Based on Big Data Analytics

利用传统算法进行的海量数据下的银行不良信贷风险评估过程中,不良信贷风险因素的影响以及不良信贷风险管理绩效评估体系的层次结构的各个指标数据不够完整会造成对银行不良信贷风险评估的准确率降低。本文提出了一种基于APH算法的大数据分析的银行不良信贷风险评估方法,将APH算法与BSC理论相结合,获得银行不良信贷风险管理绩效评估体系的层次结构。在约束条件范围内利用德尔菲法对层次结构的各个指标数据进行优化训练,确定指标权重。建立银行不良信贷风险评估模型,从而得到准确的银行不良信贷风险评估结果。实验结果表明,利用改进算法进行海量数据下的银行不良信贷风险评估,能够提高银行不良信贷风险评估的准确率,效果令人满意。

When adopting the traditional algorithm to proceed the risk assessment of bank bad credit under the massive da?ta, risk factors of bad credit and the incompleteness of various indicator data of the hierarchical structure of risk manage?ment's performance appraisal system will cause the accuracy's reduction in bank bad credit's risk assessment. Based on APH algorithm, the paper puts forward a kind of risk assessment method about bank bad credit supported by big data ana?lytics, by combining APH algorithm and BSC theory, it gets the hierarchical structure of the performance appraisal system of bank bad credit's risk assessment. Within the scope of the constraint condition, by using Delphi method to proceed the optimal training on every index data of the hierarchical structure, the index weight can be determined. And then by build?ing the risk assessment model of bank bad credit, the accurate assessment result of bank bad credit can be obtained. Ex?perimental results show that adopting the improved algorithm to proceed the risk assessment of bank bad credit under the massive data, the accuracy of risk assessment of bank bad credit can be enhanced, thus having satisfactory effect.