Predicting physical properties (viscosity, density, and refractive index) of ternary systems containing 1-octyl-3-methyl-imidazolium bis(trifluoromethylsulfonyl)imide, esters and alcohols at 298.15Â K and atmospheric pressure, using rigorous classificatio

With the use of artificial neural networks (ANNs) and gene expression programming (GEP), 362 experimental data of 3 ternary [C8mim][NTf2], ester and alcohol mixtures were mathematically modeled. The ANNs provided a robust multi-target network, suited to use as a part of a computer program, which predicted viscosity, density, and refractive index of the ternary mixtures simultaneously. The GEP models provided five separate correlations, suited to use for hand calculations, to predict viscosity, density, and refractive index separately. A comprehensive error analysis was performed on the proposed models to guarantee their accuracy. To assure quality of the experimental data used in developing the model an outlier detection was conducted using Leverage approach. At the end, by using a sensitivity analysis, the effect of each input parameter was investigated.