摘要
为了降低由评价标准可靠性较低引起的风险评估误差,提出基于改进BP神经网络的商业保险介入风险评估系统设计。主控芯片通过SPI控制器调动内置存储器内的数据,利用BP神经网络对影响风险的因素进行赋权,通过反向计算检验并调整异常赋权结果,以此为评价标准构建用户自画像,再根据画像评估商业保险的介入风险。实验结果表明,该评估系统计算出的风险结果与实际结果之间具有较高的拟合度,同时能够对主要影响因素以及风险存在形式进行准确评估。
In order to reduce the risk assessment error caused by the low reliability of the evaluation standard,a commercial insurance intervention risk assessment system design based on improved BP neural network is proposed.The main control chip mobilizes the data in the built-in memory through the SPI controller,uses the BP neural network to weight the factors that affect the risk,checks and adjusts the abnormal weighting results through reverse calculations,uses this as the evaluation standard to construct a user self-portrait,and then according to The portrait assesses the intervention risk of commercial insurance.The experimental results show that there is a high degree of fit between the risk results calculated by the design system and the actual results,and the main influencing factors and the forms of risk can be accurately evaluated.
作者
胡盈盈
桑珍珍
HU Yingying;SANG Zhenzhen(Zhengzhou Business University,Gongyi Henan 451200,China)
出处
《信息与电脑》
2021年第13期88-90,共3页
Information & Computer
关键词
改进BP神经网络
商业保险
介入风险
主控芯片
improve BP neural network
business insurance
intervention risk
master chip
