The phenomenon of vibration is prevalent in many of the engineered systems. The primary function of the suspension system of an automotive vehicle is to control the effects of vibrations and maintain stability. In a suspension system, the shock absorber (damper) acts as the primary energy absorbing element for reducing the vibrations. The performance characteristics of a damper represent highly nonlinear behavior especially in the Force-velocity characteristics. The modeling of damper involves complex nonlinear terms requiring computationally expensive resources. The presented research work utilizes three machine learning models of Recurrent Neural Networks (RNN), Long Short Term Memory model (LSTM) with single layer and Long Short Term Memory model with two layers. Out of the three machine learning models, Long Short Term Memory model with two layers is capable of accurate prediction of the damper characteristics as evident from the performance metrics of the models. Such machine learning models aid in simplifying the formulation of control algorithms that are required for semi active control of dampers.