Deep Learning-Enhanced Remote Sensing for Analysing Socio-Economic Impacts of Climate Change: An Interdisciplinary Perspective

Climate change poses significant challenges across socio-economic and environmental spectrums, necessitating advanced analytical tools to understand and mitigate its impacts. This study presents a novel Hybrid GNN-Transformer Model designed to analyse complex datasets with spatial and temporal dynamics, particularly focusing on the socio-economic impacts of climate change. Integrating the strengths of Graph Neural Networks (GNNs) and Transformers, this model adeptly processes spatial relationships and temporal sequences to provide deep insights into how climate change influences socio-economic factors across different regions and times. Employed on a dataset reflecting various socio-economic and environmental parameters, the proposed model achieves a remarkable accuracy of 96.89%. This performance not only validates the effectiveness of combining GNNs and Transformers but also underscores the potential of this approach in extracting meaningful patterns from complex, multi-dimensional data. The model’s capabilities are demonstrated through various metrics, including precision, recall, and F1-score, establishing its superiority over traditional models in handling the intricacies of integrated spatial and temporal data.