摘要
为了合理配置卫生资源,建立门诊量精确预测模型,本文使用门诊历史数据结合气象数据和环境监测数据,采用差分处理的xgboost方法进行预测。结果表明,此模型在测试集上决定系数R^2为0.805,平均绝对百分比误差mape为4.7%,优于未加入气象数据及环境监测数据的门诊量预测(R^2为0.757,mape为5.3%)。该模型能够对门诊量进行较为准确的预测,为日值卫生资源的合理分配提供依据。
In order to allocate the hospital’s health resources rationally and establish an accurate model for outpatients volume prediction,history outpatient data combined with meteorological data and environmental monitoring data is used with differential processing and xgboost methods for prediction.The results show that the determination coefficient R^2 is 0.805 and the mean absolute percentage error mape is 4.7%in test dataset.The R^2 and mape of outpatients volume prediction which did not include meteorological data and environmental monitoring data is 0.757 and5.3%.This model can predict the outpatient volume of the hospital accurately and provide a basis for allocating the daily health resources rationally.
作者
张家艳
郑建立
ZHANG Jiayan;ZHENG Jianli(University of Shanghai for Science and Technology,Shanghai 200093,China)
出处
《智能计算机与应用》
2020年第2期204-207,共4页
Intelligent Computer and Applications
关键词
气象数据
环境监测因素
差分
xgboost
门诊量
meteorological data
environmental monitoring data
difference
xgboost
outpatient visits
