摘要
为提高地下水位观测数据中干扰事件的识别效率,利用决策树算法对宝坻等5个台站近5年的水位观测数据进行样本训练和数据验证。结果表明,决策树算法对观测系统干扰和场地环境干扰事件的分类准确率在80%以上。在大量准确的训练样本基础上,决策树算法对于各种水位干扰事件具有良好的识别效果。
To improve the identification efficiency of disturbance events in groundwater observation data,decision tree algorithm is used to perform sample training and data verification for groundwater data of Baodi and other four stations in recent five years.The results show that the classification accuracy of the decision tree algorithm for observing system interference and environmental interference events is above 80%.Based on a large number of accurate training samples,the decision tree algorithm can identify various water level interference events efficiently.
作者
尹晶飞
张明
沈钰
严俊峰
徐梦林
Yin Jingfei;Zhang Ming;Shen Yu;Yan Junfeng;Xu Menglin(Zhejiang Earthquake Agency,Hangzhou 310013,China)
出处
《国际地震动态》
2019年第11期27-34,共8页
Recent Developments in World Seismology
基金
浙江省地震局局科技项目(2018zjj07)资助
关键词
水位观测
决策树
干扰识别
groundwater observation
decision tree
interference identification
作者简介
通信作者:尹晶飞,e-mail:858079816@qq.com。
