Centre de documentation
scientifique

Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling / Stefanos Georganos in Geocarto international, vol 36 n° 2 ([01/02/2021])  

Public [article]  inGeocarto international > vol 36 n° 2 [01/02/2021] . - pp 121 -1 36 Titre : Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling Type de document : Article/Communication Auteurs : Stefanos Georganos, Auteur ; Tais Grippa, Auteur ; Assane Niang Gadiaga, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : pp 121 -1 36 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] apprentissage automatique [Termes IGN] autocorrélation spatiale [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] Dakar (Sénégal) [Termes IGN] densité de population [Termes IGN] distribution spatiale [Termes IGN] hétérogénéité spatiale [Termes IGN] modèle dynamique [Termes IGN] population [Termes IGN] utilisation du sol Résumé : (auteur) Machine learning algorithms such as Random Forest (RF) are being increasingly applied on traditionally geographical topics such as population estimation. Even though RF is a well performing and generalizable algorithm, the vast majority of its implementations is still ‘aspatial’ and may not address spatial heterogenous processes. At the same time, remote sensing (RS) data which are commonly used to model population can be highly spatially heterogeneous. From this scope, we present a novel geographical implementation of RF, named Geographical Random Forest (GRF) as both a predictive and exploratory tool to model population as a function of RS covariates. GRF is a disaggregation of RF into geographical space in the form of local sub-models. From the first empirical results, we conclude that GRF can be more predictive when an appropriate spatial scale is selected to model the data, with reduced residual autocorrelation and lower Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) values. Finally, and of equal importance, GRF can be used as an effective exploratory tool to visualize the relationship between dependent and independent variables, highlighting interesting local variations and allowing for a better understanding of the processes that may be causing the observed spatial heterogeneity. Numéro de notice : A2021-080 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2019.1595177 Date de publication en ligne : 10/06/2019 En ligne : https://doi.org/10.1080/10106049.2019.1595177 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96822 [article]