基于径向基函数神经网络的口岸鼠密度动态预测  

Dynamic prediction of rat density on radial basis function neural network at port

作　　者：吴海磊[1] 钱吉生[1] 张纯[1] 陆永昌[2] 阮治安[1] 吕永生[1] 徐瑞平[1] 

机构地区：[1]南京出入境检验检疫局卫生与食品检验监督处,江苏南京211106 [2]江苏出入境检验检疫局

出　　处：《中国国境卫生检疫杂志》2012年第3期194-196,207,共4页Chinese Journal of Frontier Health and Quarantine

摘　　要：目的建立口岸鼠密度变化的动态模拟径向基函数神经网络模型,分析预测效果。方法监测鼠密度,分析鼠密度与气象因子相关性,运用多元回归方程分析气象因子对鼠密度的影响,建立鼠密度变化的动态模拟径向基函数神经网络模型,分析模型的准确性。结果建立的模型的训练准确率为91.34%,检验准确率为91.17%,测试准确率为89.03%,平均准确率为90.51%。模型认为自变量的重要性排序依次为月均最低气温、月均相对湿度、日照、降水量。结论径向基函数神经网络技术能够较好地应用到鼠密度动态预测工作中,为口岸鼠类防控提供了科学依据。Objective To construct dynamic prediction model of rat density on radial basis function neural network, and to analyze the effectiveness of this model. Methods Rat density was monitored, correlation between rat density and meterorology factors were analyzed, multiple regression equation was used to study the effect of meterorology factors on rat density, and dynamic prediction model of rat density was constructed by radial basis function neural network. Results For the constructed model, the overall forecasting accurate rate of this model for training sample and test sample were 91.34% and 91.17%. For another independent sample, overall forecasting accurate rate was 89.03%, and the average accurate rate for forecasting was 90.51%. Monthly average minimum temperature, monthly average relative humidity, sunshine and amount of precipitation were the most important influencing factors of rat density. Conclusion Radial basis function neural network could be applied to forecasting rat density for prevention and control of rats at ports.

关 键 词：鼠密度 气象因素 径向基函数神经网络 预测 

分 类 号：R311[医药卫生—基础医学]