Ionospheric scintillation forecasting and modeling are vital for efficiently tracking satellites and navigation systems. Scintillations modulate the amplitude or phase of a signal waveform caused by abnormalities of the ionospheric electron density. These fluctuating signals can cause cycle slips, disconnect the receiver signal, and cause lock loss. In the current article, we predict the amplitude of scintillation (S4 index) using a machine-learning approach. A feedforward backpropagation technique was implemented. For further learning of models regarding the dynamics of the ionospheric F layer, we inserted foF2 and hmF2 parameters in the input layer neurons. The ground–based SCINDA data at Helwan, Egypt (29.86° N, 31.32° E) from 2009 to 2017 has been considered. The results show that predicted S4 values closely reflect observed S4 values for different conditions of the solar cycle 24, with a RMSE of 0.019 and regression of 0.659. The variations of ionospheric scintillation near the equatorial anomaly's northern peak have also been conducted during different levels of solar cycle 24 based on the ANN.