In this communication, the solubility of sulfur dioxide in various ionic liquids is estimated using the least square support vector machine (LSSVM) combined with group contribution method. A dataset comprised of 232 data points on SO2 solubility in different ionic liquids was established to develop the LSSVM model. The proposed model used the temperature, pressure, and 17 chemical structures for ionic liquids as input parameters. The hybrid LSSVM was trained using 75% of data points while the other 25% were considered as a testing dataset. It was found that the promising results were obtained by LSSVM model parameters of gamma = 25436.514 and sigma(2) = 1.0365 using an optimization procedure by genetic algorithm (GA). Coefficient of determination (R-2) and percentage of absolute average relative deviation (%AARD) are 0.9978 and 1.42%, respectively, for the proposed hybrid LSSVM model. (C) 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.