Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexity

Mapping environmental variables is crucial for natural resource management. Researchers and scholars have continually advanced this field with modern techniques such as Integrated Nested Laplace Approximation (INLA), Deep Learning (DL), and Graph Neural Networks (GNN) models. While effective, these...

Full description

Saved in:
Bibliographic Details
Main Authors: Emanuele Barca, Maria Clementina Caputo, Rita Masciale
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:International Journal of Applied Earth Observations and Geoinformation
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1569843225001256
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849322043577729024
author Emanuele Barca
Maria Clementina Caputo
Rita Masciale
author_facet Emanuele Barca
Maria Clementina Caputo
Rita Masciale
author_sort Emanuele Barca
collection DOAJ
description Mapping environmental variables is crucial for natural resource management. Researchers and scholars have continually advanced this field with modern techniques such as Integrated Nested Laplace Approximation (INLA), Deep Learning (DL), and Graph Neural Networks (GNN) models. While effective, these models often present a significant challenge due to their black nature, which obscures the process of generating final maps from raw data. Recent theoretical breakthroughs have shown that white/grey-box models can achieve the same level of accuracy as these advanced techniques, debunking the belief that complex models are necessarily the most accurate. Based on these findings, we have developed a methodology that employs a series of statistical tests and data analytics to identify essential features hidden in spatial data in order to assess the predictive model (of white/grey kind) that best approximates underlying spatial processes. This methodology profiles the model that better adapts to the data, aiding in the selection of the simplest model that achieves the desired accuracy, functioning similarly to a recommender system for model selection. Furthermore, the set of permissible models includes only regressive-like ones to clarify the data’s contribution to map construction and can be applied to a wide range of datasets. By reducing complexity, this approach enhances the transparency of the model’s results. Real-world dataset demonstrates this methodology’s remarkable ability to produce highly accurate results.
format Article
id doaj-art-6273fc25e30546d3ad04a97b4a3c755f
institution Kabale University
issn 1569-8432
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series International Journal of Applied Earth Observations and Geoinformation
spelling doaj-art-6273fc25e30546d3ad04a97b4a3c755f2025-08-20T03:49:33ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322025-05-0113910447810.1016/j.jag.2025.104478Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexityEmanuele Barca0Maria Clementina Caputo1Rita Masciale2Corresponding author.; Water Research Institute, National Research Council of Italy, 70132 Bari, ItalyWater Research Institute, National Research Council of Italy, 70132 Bari, ItalyWater Research Institute, National Research Council of Italy, 70132 Bari, ItalyMapping environmental variables is crucial for natural resource management. Researchers and scholars have continually advanced this field with modern techniques such as Integrated Nested Laplace Approximation (INLA), Deep Learning (DL), and Graph Neural Networks (GNN) models. While effective, these models often present a significant challenge due to their black nature, which obscures the process of generating final maps from raw data. Recent theoretical breakthroughs have shown that white/grey-box models can achieve the same level of accuracy as these advanced techniques, debunking the belief that complex models are necessarily the most accurate. Based on these findings, we have developed a methodology that employs a series of statistical tests and data analytics to identify essential features hidden in spatial data in order to assess the predictive model (of white/grey kind) that best approximates underlying spatial processes. This methodology profiles the model that better adapts to the data, aiding in the selection of the simplest model that achieves the desired accuracy, functioning similarly to a recommender system for model selection. Furthermore, the set of permissible models includes only regressive-like ones to clarify the data’s contribution to map construction and can be applied to a wide range of datasets. By reducing complexity, this approach enhances the transparency of the model’s results. Real-world dataset demonstrates this methodology’s remarkable ability to produce highly accurate results.http://www.sciencedirect.com/science/article/pii/S1569843225001256Fundamental spatial assumptionSpatial trendStationary stochastic modelHybrid modelRecommender system
spellingShingle Emanuele Barca
Maria Clementina Caputo
Rita Masciale
Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexity
International Journal of Applied Earth Observations and Geoinformation
Fundamental spatial assumption
Spatial trend
Stationary stochastic model
Hybrid model
Recommender system
title Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexity
title_full Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexity
title_fullStr Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexity
title_full_unstemmed Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexity
title_short Building the optimal hybrid spatial Data-Driven Model: Balancing accuracy and complexity
title_sort building the optimal hybrid spatial data driven model balancing accuracy and complexity
topic Fundamental spatial assumption
Spatial trend
Stationary stochastic model
Hybrid model
Recommender system
url http://www.sciencedirect.com/science/article/pii/S1569843225001256
work_keys_str_mv AT emanuelebarca buildingtheoptimalhybridspatialdatadrivenmodelbalancingaccuracyandcomplexity
AT mariaclementinacaputo buildingtheoptimalhybridspatialdatadrivenmodelbalancingaccuracyandcomplexity
AT ritamasciale buildingtheoptimalhybridspatialdatadrivenmodelbalancingaccuracyandcomplexity