Among several available lightweight materials, different aluminium alloys (AA), especially the aerospace grades, have gained popularity in automobile and aerospace industries, owing to their high strength-to-weight ratio. But joining of AA through conventional fusion techniques is difficult due to the formation of oxide layers, intermetallic compounds, etc. These issues have been successfully addressed through eliminating the melting of metals employing a solid-state friction stir welding (FSW) process. Although some literature on FSW of different Al grades is available, but studies on the influence of FSW parameters on the AA2024-AA7075 joint strength is limited. The present study thus addresses this gap by developing a cubic statistical regression analysis (SRA), single-layer artificial neural network (SL-ANN) and multiple layer(s) complex neural network (MLCNN) models. All the models have predicted with reasonable accuracy, where MLCNN has been observed to perform better than SL-ANN owing to its complex architecture and rigorous data processing for error minimization. SRA has been observed to outperform all other algorithms.