We describe an artificial neural network model of the near-Earth space radiation environment. The geomagnetic activity index  K_p and low-earth-orbit (LEO) electron flux measurements from the National Oceanic and Atmospheric Administration Polar-orbiting Operational Environmental Satellite (POES) operational spacecraft are used as model training inputs. Electron fluxes from National Aeronautics and Space Administration's Van Allen Probe spacecraft form the training outputs. We demonstrate that the model can accurately specify outer radiation belt (L ∼3-7) electron fluxes at two energies, 350 keV and 1 MeV. Various performance metrics are calculated using out-of-sample data, and we find high correlations and low errors between the model specification and the observed flux. We emphasize that once the model is trained, the Van Allen Probes data are no longer needed at model run time; only the POES fluxes and the K_p index are required to specify the outer electron belt using the model.