摘要
目的探索BP人工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,建立福州地区菌痢流行的BP人工神经网络模型,并以2007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0.826,模型的预测成功率为83.33%。结论本研究表明,BP人工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 the performance of a BP neural network model of bacillary dysentery prevalence influenced by meteorological elements,and to provide scientific evidences for preventive strategies of bacillary dysentery.Methods The forecasting model of bacillary dysentery prevalence was established by using the neural network toolbox of Matlb 7.2 software package with surveillance data of bacillary dysentery and meteorological records from the year of 1987 to 2006 in Fuzhou.The model was tested with data of 2007.Results The training of the model decreased errors of predictions and made the system more stable.The coefficient value of final model was 0.826 and the accurate rate of prediction was 83.33%.Conclusion The established BP neural network model may be used as a new method for predicting the prevalence of bacillary dysentery with meteorological elements.
出处
《中国预防医学杂志》
CAS
2010年第2期178-180,共3页
Chinese Preventive Medicine
关键词
细菌性痢疾
气象要素
BP人工神经网络
Bacillary dysentery
Meteorological elements
BP neural network
