应用BP神经网络模型预测福州市山区细菌性痢疾流行  

PREDICTION OF BACILLARY DYSENTERY BY BP NEURAL NETWORK MODEL IN MOUN TAINOUS AREA IN FUZHOU

作　　者：沈波[1] 王李仁[1] 许旭艳[1] 郑能雄[1] 

机构地区：[1]福州市疾病预防控制中心,福州350014

出　　处：《现代预防医学》2011年第3期423-424,429,共3页Modern Preventive Medicine

摘　　要：[目的]探索BP神经网络在细菌性痢疾预测模型的应用,为细菌性痢疾的预防控制措施提供科学依据。[方法]用Matlab7.2软件包中的神经网络工具箱,以1988～2007年的资料建立福州市山区菌痢流行的BP神经网络模型,并以2008年的资料验证其预测成功率。[结果]神经网络经学习和训练,训练误差下降并趋于稳定,回代相关系数为0.815,模型的预测成功率为10/12。[结论]BP神经网络在气象要素与菌痢发病之间建模是可行的,可以作为预测菌痢流行的一种新方法。[Objective] To explore application of BP neural network model in prediction of bacillary dysentery, in order to provide the scientific data for making strategies. [Methods] The forecasting model for bacillary dysentery was established by using the neural network toolbox of Matlab7.2 software package. In the studies of forecasting model, the data in Fuzhou from 1988 to 2007 were chosen to analyze. The established forecasting model was also tested by the data of bacillary dysentery in 2008. [Results] After training the neural network, the error of performance decreased and the coefficient of regression was 0.815. The efficiency of the forecasting model for bacillary dysentery was 10 / 12. [Conclusion] BP neural network model is feasible to analyze the relation of meteorological factors and bacillary dysentery. BP neural network model could be used as a new effective method for forecasting of bacillary dysentery.

关 键 词：细菌性痢疾 气象要素 BP人工神经网络 

分 类 号：R516.4[医药卫生—内科学]