基于支持向量机的麻疹发病率预测研究  

Prediction of measles incidence rate based on the support vector machine model

作　　者：徐学琴[1] 裴兰英[1] 王瑾瑾[1] 刘晓蕙[1] 孙春阳[1] 闫国立[1] 

机构地区：[1]河南中医药大学基础医学院,河南郑州450046

出　　处：《中华疾病控制杂志》2017年第5期528-530,共3页Chinese Journal of Disease Control & Prevention

基　　金：河南省软科学研究重点项目资助(102400440002);河南中医学院"科研苗圃工程项目"(MP2014-07)

摘　　要：目的建立一种基于支持向量机的麻疹发病率预测模型,为麻疹的预防控制决策提供参考依据。方法收集我国麻疹1996-2015年发病率数据,以1996-2014年发病率数据为训练样本,以2015年发病率数据为检验样本,采用支持向量机回归算法建立我国麻疹年发病率的预测模型,并对我国2016-2018年的麻疹发病率进行预测。结果经该模型的预测,麻疹发病率的预测值和实际值吻合度较高,平均相对误差为0.620 07%。我国2016-2018年麻疹的发病率分别为3.23/10万、3.13/10万、3.79/10万。结论采用支持向量机回归模型预测我国麻疹的年发病率是可行有效的。Objective To establish a support vector machine model for prediction of measles incidence rate,in order to provide the reference for measles prevention and control decision.Methods The data of measles incidence rate in china from 1996 to 2015 were collected.The incidence rates of measles from 1996 to 2014 were training samples,and the incidence rate of 2015 was testing sample.The prediction model was established based on the support vector machine regression algorithm,and the incidence rates of measles from 2016 to 2018 were predicted by using this model.Results The actual incidence rates and predicted incidence rates of measles were highly consistent;the average relative error was 0.620 07%.The predicted incidence rates of measles in china from 2016 to 2018 were 3.23/100 000 and 3.13/100 000,3.79/ 100 000,respectively.Conclusions It is feasible and effective to predict the incidence rate of measles by using the support vector machine regression model.

关 键 词：麻疹 发病率 预测 

分 类 号：R272.22[医药卫生—中医儿科学]