Interactive web-based Geospatial eXplainable Artificial Intelligence for AI model output exploration

Safariallahkheili, Qasem; Schiewe, Jochen; Meier, Sebastian

doi:10.5194/agile-giss-6-44-2025

Article / Chapter Mon Jun 9 2025 CC BY 4.0

published

Interactive web-based Geospatial eXplainable Artificial Intelligence for AI model output exploration

Safariallahkheili, Qasem ; Schiewe, Jochen ; Meier, Sebastian

This case study presents a web-based Geospatial eXplainable Artificial Intelligence (GeoXAI) system demonstrated through a case study for wildfire susceptibility assessment. Addressing limitations in traditional GeoXAI tools, the system integrates XAI methods with open-source geospatial technologies. Using a Random Forest model, the system combines environmental, topographic, and meteorological features to provide global and local insights. SHAP values offer feature-level explanations, while the interactive platform enables users to visualize wildfire susceptibility, examine feature contributions, and correlate predictions with spatial patterns and distribution of feature values. This approach tries to enhance transparency in AI-driven environmental decision support systems, with a specific focus on the interpretability of model output.

Preview

Category

Conference:: 28th AGILE Conference on Geographic Information Science (Veranstaltung : 2025 : Dresden)
Published in:: AGILE: GIScience series
Vol. 6.2025 (09.06.2025), art. no. 44, p. 1-7
Volume:: 6.2025
Publication date:: 09.06.2025
URN:: urn:nbn:de:0307-20260309-002-9
DOI:: 10.5194/agile-giss-6-44-2025
Language:: English
Type of Resource:: Text
Extent:: 1 Online-Ressource (Seite 1-7, 9,79 MB) : 3 Illustrationen, 1 Textdatei (PDF)
Keywords:: wildfire susceptibility; explainable artificial intelligence; GeoXAI; GIS; Random Forest^GND; Anfälligkeit für Waldbrände; Erklärbare künstliche Intelligenz^GND; Geoinformationssystem^GND
DDC subject of DNB:: 520 Astronomy & allied sciences
URL:: https://doi.org/10.5194/agile-giss-6-44-2025
URL:: https://agile-giss.copernicus.org/articles/6/44/2025/
Link to Raw Object:: https://agile-giss.copernicus.org/articles/6/44/2025/agile-giss-6-44-2025.pdf
Hosting institution:: University of Applied Sciences Potsdam (FHP) [DE]
Hosting institution:: HafenCity University Hamburg (HCU) [DE]
Statement of responsibility:: Qasem Safariallahkheili, Jochen Schiewe, and Sebastian Meier
Bibliographic notes:: Published in: AGILE: GIScience Series, 6, 2025 - 28th AGILE Conference on Geographic Information Science “Geographic Information Science responding to Global Challenges”, 10–13 June 2025, Dresden, Germany. Editors: Auriol Degbelo, Serena Coetzee, Carsten Keßler, Monika Sester, Sabine Timpf, and Lars Bernard