Owning to amazing nature of ionic liquids (ILs), they can be applied in numerous chemical fields and industrial plants. ILs have great potential to capture greenhouse gases. On the other hands, experimental measurements are time-consuming and costly. Hence, looking for a precise model to estimate the solubility of CO2 as an important greenhouse gas in the presence or absence of impurities like H2S, SO2, CH4, H-2, and N2O and different solvents like ILs and Amines is certainly crucial. To this end, Least Square Support Vector Machine (LSSVM) algorithm was developed as a reliable tool for estimating the solubility of CO2-rich gaseous mixtures containing H2S, SO2, CH4, and N2O and various ionic liquids like [C8mim] [PF6], [C8mim] [Tf2N], [bmim] [MeSO4], [bmim] [PF6], [bmim] [Tf2N], [emim] [dep], [thtdp] [dca], [thtdp] [phos], [hmim][Tf2N], [bmim] [BF4], and [bmim] [Ac] as function of temperature, pressure, mole fraction of ILs, mole ratio of CO2 to second gas in feed, and physical properties of ILs and second gas including molecular weight, critical pressure, and critical temperature. In addition, Genetic Algorithm (GA) was applied to determine the hyperparameters of the LSSVM (sigma(2) and gamma). Results obtained from the statistical analyses confirm this fact that the approximations are in good agreement with the actual reported data points. A comparison was carried out between the obtained results from the LSSVM model and three different kinds of machine learning approaches such as the multilayer perceptron artificial neural network (MLP-ANN), radial basis function artificial neural network (RBF-ANN), and adaptive network-based fuzzy inference system (ANFIS). The statistical analyses such as the Mean Relative Error (MRE%) and R-Square (R-2) were obtained as 0.7545 and 1.0000, 18.06 and 0.9956, 7.016 and 0.9993, 50.63, 0.9628 for the LSSVM, RBF-ANN, MLP-ANN, and ANFIS, respectively. The efforts in this study cover the way for solubility estimations of carbon dioxide in ternary mixtures, which can help chemists and engineers to have a reliable tool with low dependent parameters for monitoring the conditions and phase behavior of the systems.