A geographically weighted artificial neural network

While recent developments have extended geographically weighted regression (GWR) in many directions, it is usually assumed that the relationships between dependent and independent variables are linear. In practice, however, often case nonlinearly associated. To address this issue, we propose a artificial neural network (GWANN). GWANN combines geographical weighting with networks, which able to learn complex nonlinear data-driven manner without assumptions. Using synthetic data known spatial characteristics real-world study, compared GWR. results for show performs better than GWR when within their variance high, based on demonstrate performance of can also be superior practical setting.