Reservoir fluid is one of the major parameters that play a significant role in oil mobility, the fluid flow through the porous of the reservoir, the mechanism at which this fluid has been migrated are some of the major concerns in enhanced oil recovery industries. So, it of great significance to use an appropriate technique to calculate or to predict the oil recovery from the reservoir. The study aims to use an artificial neural network (ANN) model with electromagnetic parameters such as real and imaginary permittivity, real and imaginary permeability, and reflection loos at different concentrations of NaCl electrolyte. Deep neural network (DNN) approach with 280 nodes in each hidden layer and one output, it was revealed from the obtained result the correlation coefficient given by the DNN is R2 of 0.994 for estimated recovery factor and R2 of 0.894 for predicted recovery factor. This research result shows a good prediction of RF with the reservoir rock and fluid properties in terms of cost and production effectiveness. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.