基于BP神经网络的资产经营绩效评估模型  

Performance Evaluation Model of Capital Operation about Industrial Enterprises Based on Optimized BP Neural Network

作　　者：陈当阳[1] 贾素玲[1] 王惠文[1] 罗昌[1] 

机构地区：[1]北京航空航天大学经济管理学院,北京100083

出　　处：《北京航空航天大学学报（社会科学版）》2007年第3期1-5,共5页Journal of Beijing University of Aeronautics and Astronautics：Social Sciences edition Edition

基　　金：国家自然科学基金创新研究群体科学基金资助项目(70521001)

摘　　要：针对传统资产经营绩效评估模型的缺陷,提出使用BP神经网络对企业的资产经营进行绩效评估,并对传统的BP网络进行了优化。其中自适应学习机制优化了学习率,动量项加速机制加速了传统BP网络的收敛速度,遗传算法用于选择BP网络的初始参数。最后,采用500家工业企业的资产经营绩效数据对模型的有效性进行了初步验证。According to defects of traditional performance evaluation models, BP neural network is used to evaluate performance of capital operation. Some optimizations to BP neural networks are done. The learning mechanism of self adaptation optimizes the learning ratio of BP neural networks. The accelerating mechanism of momentum item accelerates the constringent velocities of BP neural networks. The genetic algorithm is used to select initial parameters of BP neural networks. Lastly, performance data of capital operation in 500 industrial enterprises is used to prove validity of the model.

关 键 词：资产经营 指标体系 BP神经网络 遗传算法 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]