CAJ | 학술논문

提出一种基于递阶遗传算法和BP神经网络的财务预警模型.现有的BP网络模式分类训练方法大都只能训练BP网络的权重,网络的结构得预先用某种方法确定.利用巧妙设计的递阶遗传算法能够把网络的结构和权重同时通过训练确定.以模式分类数据库中的数据进行训练和测试,并与其他模式分类模型相比较.结果表明,该模型更优,分类精确度更令人满意.根据上市公司的财务数据用所提出的方法进行财务预警是可行的.
제출일충기우체계유전산법화BP신경망락적재무예경모형.현유적BP망락모식분류훈련방법대도지능훈련BP망락적권중,망락적결구득예선용모충방법학정.이용교묘설계적체계유전산법능구파망락적결구화권중동시통과훈련학정.이모식분류수거고중적수거진행훈련화측시,병여기타모식분류모형상비교.결과표명,해모형경우,분류정학도경령인만의.근거상시공사적재무수거용소제출적방법진행재무예경시가행적.
A financial crisis warning model, based on hierarchical genetic algorithm and BP neural network, is proposed. Different from the existing BP training method resulting only in determination of connection weights, a well-designed hierarchical genetic algorithm is used to train BP neural network with both connection weights and numbers of neurons in a hidden layer determined at the same time. The model is then used to classify the data of Wine. It is shown that the model based on hierarchical genetic algorithm and BP neural network is simple and effective. According to the financial data of listed companies, using the method proposed to warn the finance problem is feasible.