CRTS (China railway track system)-Ⅲ slab track is a main structural type of ballastless track in Chinse high-speed railway (HSR), which requires precise positioning of prefabricated concrete track slabs. However, the cast-in-place process of the self-compacting concrete (SCC) filling layer induces a large deformation as the flowing SCC generates an up-lift force on the bottom of the track slab. Currently, the deformation mechanism and influencing factors of the track slab during the SCC casting process have not been thoroughly studied. In this work, a fluid-solid interaction finite element model (FEM) is created to calculate the upward deformation of the CRTS-Ⅲ track slab caused by the flowing SCC. The main influencing factors, including the SCC’s casting height and speed, SCC’s viscosity, constraint stiffness of the limiting beams, and vertical temperature gradient, have been systematically analyzed and discussed. The relationship between the influencing factors and the track slab’s deformation is obtained. Furthermore, machine learning prediction models using random forest algorithms are developed to evaluate the deformation and up-lift force. A dataset comprising 603 calculation cases obtained from FEM simulations is used to train the prediction models. The random forest-based models developed in this study give an accurate estimation of the deformation and up-lift force based on the SCC properties and construction parameters. In addition, the feature importance of each influencing factor is obtained and studied. The findings of this study can provide guidance for the control of the deformation in CRTS-Ⅲ track slab during the SCC construction process.