| 混沌风险探测模型的研究与应用 |
| |
| 引用本文： | 朱艳丽,李士勇.混沌风险探测模型的研究与应用[J].河南职业技术师范学院学报,2011(2):79-82. |
| |
| 作者姓名： | 朱艳丽 李士勇 |
| |
| 作者单位： | 河南科技学院,河南新乡453003 |
| |
| 基金项目： | 河南省教育厅自然科学计划（2009A520013） |
| |
| 摘 要： | BP神经网络是应用最广泛的预测模型,它能方便、灵活地对信用卡消费行为进行探测,但BP网络有很多固有缺陷,比如结构难确定、初始权值选择盲目性导致训练速度慢等,结合信用卡交易数据的混沌特征分析,通过应用混沌理论中的相空间重构技术,把信用卡客户的相关数据嵌入到重构的相空间中,然后利用BP神经网络技术建立混沌风险探测模型,对信用卡交易行为进行风险预测.实验结果表明,该模型的预测精度高于一般的神经网络预测方法,其中正确检出率比使用BP神经网络模型提高了3%.
|
| 关 键 词： | 混沌风险探测模型 混沌理论 神经网络 预测 |
Study and application of chaos-risk prediction model |
| |
| Zhu Yanli,Li Shiyong.Study and application of chaos-risk prediction model[J].Journal of Henan Vocation-Technical Teachers College,2011(2):79-82. |
| |
| Authors: | Zhu Yanli Li Shiyong |
| |
| Institution: | (Henan Insitute of Science and Technology,Xinxiang 453003,China) |
| |
| Abstract: | BP neural network is the most widely used forecasting model that can easily and flexibly detect the credit card consumer behavior,but BP network has many inherent shortcomings,such as difficult to determine structure,initial weights chosen blindnessly etc.This paper has analyzed the chaos property of credit card transaction data,through application of chaos theory in the phase space reconstruction,and then a new forecasting model based on Chaos theory and Neural Network is developed to predict the risk of credit card transactions.Experiments show that true positive prediction rate of the proposed model is 3% higher than neural network prediction method. |
| |
| Keywords: | chaos-risk prediction model neural network chaos theory prediction |
| 本文献已被 维普 等数据库收录！ |
|
