Locally varying geostatistical machine learning for spatial prediction

作     者：Francky Fouedjio Emet Arya 

作者机构：Rio TintoData&Analytics152-158 St Georges TerracePerthWA 6000Australia Kaplan Business School Pty LtdPerth Campus1325 Hay StWest PerthWA6005Australia Edith Cowan UniversitySchool of Science270 Joondalup DriveJoondalupWA 6027Australia 

出 版 物：《Artificial Intelligence in Geosciences》 (地学人工智能（英文）)

年 卷 期：2024年第5卷第1期

页      面：28-45页

核心收录：

学科分类：0502[文学-外国语言文学] 050201[文学-英语语言文学] 05[文学] 

基　　金：The authors are grateful to anonymous reviewers and the associate editor for their helpful and constructive comments that helped improve the manuscript 

主　　题：Data augmentation Geostatistics Local stationarity Machine learning Conditional simulation Spatial auto-correlation Spatial non-stationarity Spatial uncertainty 

摘      要：Machine learning methods dealing with the spatial auto-correlation of the response variable have garnered significant attention in the context of spatial ***,under these methods,the relationship between the response variable and explanatory variables is assumed to be homogeneous throughout the entire study *** assumption,known as spatial stationarity,is very questionable in real-world situations due to the influence of contextual ***,allowing the relationship between the target variable and predictor variables to vary spatially within the study region is more ***,existing machine learning techniques accounting for the spatially varying relationship between the dependent variable and the predictor variables do not capture the spatial auto-correlation of the dependent variable ***,under these techniques,local machine learning models are effectively built using only fewer observations,which can lead to well-known issues such as over-fitting and the curse of *** paper introduces a novel geostatistical machine learning approach where both the spatial auto-correlation of the response variable and the spatial non-stationarity of the regression relationship between the response and predictor variables are explicitly *** basic idea consists of relying on the local stationarity assumption to build a collection of local machine learning models while leveraging on the local spatial auto-correlation of the response variable to locally augment the training *** proposed method’s effectiveness is showcased via experiments conducted on synthetic spatial data with known characteristics as well as real-world spatial *** the synthetic(***)case study,the proposed method’s predictive accuracy,as indicated by the Root Mean Square Error(RMSE)on the test set,is 17%(resp.7%)better than that of popular machine learning methods dealing with the response variable’s spatial auto-correla