The effect of electromagnetic fields on evaporation rate is investigated using a hybridised prediction method based on high-fidelity computational methods and deep learning. For this objective, the volume of fluid approach in the multiphase model and turbulent flow (2,630<Re<21,203) is analysed in the presence of a constant magnetic field. The results of a computational method are used as training data for a feed-forward neural network after careful verification. Temperature, magnetic field, component velocity and vorticity, and vapour concentration are assumed as inputs in this supervised deep learning method, while convection heat transfer coefficient and evaporation rate are the targets. The present study uses five hidden layers and thirty-two learnable neurons to demonstrate output behaviour. According to the results, this technique can reduce computational costs by up to 18% compared to conventional multiphase modelling. Additionally, applying an electromagnetic field increased the evaporation rate by 4.85 %. We found that the maximum voltage range (V = 20 kV) could increase the liquid evaporation rate by up to 15.36%.