两种不同算法在青岛市成蚊密度预测中的应用

Application of two different algorithms in prediction of adult mosquito density in Qingdao,China

目的应用多元线性回归算法和基因表达式编程算法模型预测青岛市成蚊密度,探讨其在成蚊密度预测中的可行性。方法收集青岛市2016-2019年3-11月的月最高气温、月最低气温、月平均气温、月平均湿度、月累计降水量和月累计降水天数等气象资料和月成蚊密度资料,将2016-2018年数据资料作为训练集分别用2种方法建立预测模型,以2019年成蚊密度数据作为测试集分别验证2种模型的预测性能。结果多元线性回归算法模型中训练集和测试集的相关系数分别为0.94和0.93。基因表达式编程算法模型中训练集和测试集的相关系数分别为0.97和0.96。结论基于气象资料建立的2种算法模型均可较好地预测青岛市成蚊密度,为将来开展防蚊灭蚊工作提供了数据支持。

Objective To predict the density of adult mosquito in Qingdao, China using multiple linear regression algorithm and gene expression programming algorithm models, and to investigate their feasibility in predicting the density of adult mosquito density. Methods Meteorological data(monthly maximum temperature, monthly minimum temperature, monthly mean temperature, monthly mean humidity, monthly accumulated precipitation, and monthly accumulated precipitation days) and monthly adult mosquito density from March to November, 2016-2019, in Qingdao were collected. Prediction models were built using the above two methods, respectively, with the data from 2016 to 2018 used as the training set and the adult mosquito density data in 2019 used as the test set to validate the prediction performance of the two models. Results The correlation coefficients of the training set and the test set were 0.94 and 0.93,respectively, in the multiple linear regression algorithm model, and were 0.97 and 0.96, respectively, in the gene expression programming algorithm model. Conclusion The two models based on the meteorological data can favorably predict the adult mosquito density in Qingdao, which provides data support for mosquito prevention and control in the future.