摘要
在分析影响电力客户企业公司财务状况主要因素的基础上,将一种新型特征映射网络—CPN网络理论应用于电力客户信用风险预测,建立基于CPN网络电力客户信用风险预测模型.采用实际数据对模型进行验证,并将其与LVQ网络模型、BP网络模型和传统模型相比较,证明了基于CPN网络电力客户信用风险预测模型具有较高的精度和较强的实用性.
In this paper,the CPN neural network theory is applied to forecast Electricity Customer Credit Risk after analyzing the major factors affecting the Electricity Customer Credit Risk,and a Electricity Customer Credit Risk forecasting model based on CPN neural network is established.The proposed model is then verified by using actual data and is compared with LVQ neural network,BP neural network and the traditional statistic model.The model is of high precision and great applicability.
出处
《佳木斯大学学报(自然科学版)》
CAS
2008年第5期636-638,641,共4页
Journal of Jiamusi University:Natural Science Edition
基金
咸宁学院校级项目资助.(KY0719)