文章摘要
黎健,顾君忠,毛盛华,肖文佳,金汇明,郑雅旭,王永明,胡家瑜.BP人工神经网络模型在上海市感染性腹泻日发病例数预测中的应用[J].中华流行病学杂志,2013,34(12):1198-1202
BP人工神经网络模型在上海市感染性腹泻日发病例数预测中的应用
Preliminary application of Back-Propagation artificial neural network model on the predictionof infectious diarrhea incidence in Shanghai
收稿日期:2013-07-15  出版日期:2014-10-16
DOI:10.3760/cmaj.issn.0254-6450.2013.012.010
中文关键词: 感染性腹泻  气象因素  BP人工神经网络
英文关键词: Infectious diarrhea  Meteorological factors  Back-Propagation artificial neuralnetwork
基金项目:上海市公共卫生重点学科计划(12GWZX0101)
作者单位E-mail
黎健 200336 上海市疾病预防控制中心  
顾君忠 华东师范大学计算机应用研究所  
毛盛华 200336 上海市疾病预防控制中心  
肖文佳 200336 上海市疾病预防控制中心  
金汇明 200336 上海市疾病预防控制中心  
郑雅旭 200336 上海市疾病预防控制中心  
王永明 华东师范大学计算机应用研究所 ymwang@ica.stc.sh.cn 
胡家瑜 200336 上海市疾病预防控制中心 jyhu@scdc.sh.cn 
摘要点击次数: 2305
全文下载次数: 936
中文摘要:
      目的 建立基于气象因素的上海市感染性腹泻逐H发病例数BP人工神经网络预测模型。方法 收集上海市2005-2008年感染性腹泻逐日发病例数与同期气象资料包括气温、相对湿度、降雨量、气压、日照时数、风速,通过SPearman相关分析选出与感染性腹泻相关的气象因素,用主成分分析(PCA)去除气象因素间的共线性影响。利用MatLabR2012b软件的神经网络工具箱建立感染性腹泻日发病例数的BP神经网络预测模型,并对拟合效果、外推预测能力和等级预报效果进行评价。结果 SPearman相关性分析显示,日感染性腹泻与前两天的日最高气温、最低气温、平均气温、最低相对湿度、平均相对湿度呈正相关(PP<0.01)。输入PCA提取的4个气象主成分构建BP神经网络预测模型,训练和预测样本平均绝对误差、均方根误差、相关系数、决定系数分别为4.7811、6.8921、0.7918、0.8418和5.8163、7.8062、0。7202、0.8180。模型预测值对2008年实际发病数的年平均误差率为5,30%,对感染性腹泻的等级预报正确率为95.63%+H26。结论 温度和气压对感染性腹泻日发病例数影响较大。BP神经网络模型的拟合及预测误差较小,预报正确率较高,预报效果理想。
英文摘要:
      Objective To establish BP artificial neural network predicting model regardingthe daily cases of infectious diarrhea in Shanghai.Methods Data regarding both the incidence oinfectious diarrhea from 2005 to 2008 in Shanghai and meteorological factors including temperature.relative humidity.rainfall,atmospheric pressure.duration of sunshine and wind speed within the sameperiods were collected and analyzed with the MatLab R20 l 2b software.Meteorological factors thatwere correlated with infectious diarrhea were screened by Spearman correlation analysis.Principalcomponent analysis(PCA)was used to remove the multi-colinearities between meteorological factors.Back-Propagation(BP)neural network was employed to establish related prediction models regardingthe daily infectious diarrhea incidence.using artificial neural networks toolbox.The establishedmodels were evaluated through the fitting,predicting and forecasting processesResults Data fromSpearman correlation analysis indicated that the incidence of infectious diarrhea had a highly positivecorrelation with factors as daily maximum temperature,minimum temperature,average temperature,minimum relative humidity and average relative humidity in the previous two days(P<0.01),and arelatively high negative correlation with the daily average air pressure in the previous two days(P<0.0 1).Factors as mean absolute error,root mean square error,correlation coemcient(r),and thecoefficient of determination(r2)of BP neural network model were established under the input of 4meteorological principal components.extracted by PCA and used for training and prediction.Thenappeared to be 4.7811,6.8921,0.7918,0.8418 and 5.8163,7.8062,0.7202,0.8180,respectively.Therate on mean error regarding the predictive value to actual incidence in 2008 was 5.30%and theforecasting precision reached 95.63%.Conclusion Temperature and air pressure showed importantimpact on the incidence of infectious diarrhea.The BP neural network model had the advantages of10W simulation forecasting errors and high forecasting hit rate that could ideally predict and forecastthe effects OH the incidence of infectious diarrhea.
查看全文   Html全文     查看/发表评论  下载PDF阅读器
关闭