An automated RF sensor system for accurate determination of complex permittivity of thin samples and dielectric sheets is proposed. The proposed system is computationally intelligent making use of the machine learning algorithm along with the artificial neural network (ANN) architecture and employs a coplanar waveguide sensor for the measurement of scattering coefficients of test specimens in order to obtain their dielectric properties. Different heuristics are followed for the design of the ANN-based system. The applicability of the proposed sensor system for practical usage is increased by taking into account the effect of possible air gap between the device and test specimen. This is facilitated by developing a multilayered analytical model, and combining it with the ANN-based system. The complete procedure, comprising of ANN algorithms and the analytical formulation, is implemented in MATLAB. The proposed automated standalone sensor system is validated by testing a number of standard samples in the designated frequency bands.