AEST-GTE: An Adaptive Explainable Spatial–Temporal Graph Transformer Ensemble Framework for Intelligent Flood Prediction and Real-Time Flood Susceptibility Assessment
Keywords:
Spatial–Temporal Learning, Transformer, Explainable Artificial Intelligence (XAI), Flood Prediction, Neural Network.Abstract
Flood prediction is getting increasingly complicated because of the complex interactionsbetween the hydrological, meteorological and geographical variables and the dynamic climatechanges. Conventional methods used for flood forecasting rarely have the modelling capabilityto account for a large number of environmental data sources, and for nonlinear spatial–temporal relationships. In this paper, an Adaptive Explainable
References
[1] Li, Xiang, et al. "Flood Susceptibility Mapping Using CatBoost and Geospatial Analysis." Environmental Modelling & Software, vol. 145, 2022, pp. 105120.
[2] Lin, Wen-Hao, et al. "Geospatial and Remote Sensing Techniques in Flood Mapping: A Machine Learning Perspective." Journal of Environmental Informatics, vol. 41, no. 1, 2022, pp. 24-40.
