Associate surplus substances presence in natural gases like carbon dioxide (CO2) causes prominent problems in transporting and storage stages. Unique features of ionic liquids such as low vapor pressure, excellent thermal and chemical stability, high power of dissolution, etc., have made them as green solvents and organic solvents substitution in the separation processes, especially the separation of CO2. Although many experimental works have been done to determine the ability of ionic liquids in separation of CO2, expensive and timeconsuming laboratory methods have led to a strong interest in the modeling methods. In this work, quantitative structure property relationship (QSPR) models for predicting the Henry's law constant (HLC) for CO2 dissolution in 32 ionic liquids have been developed. Chosen descriptors by genetic algorithm were used to develop two models by MLR and LSSVM methods. The three-parameter model has been considered as the main model and a detailed physical interpretation has been mentioned for all parameters. Comparison of the predicted values of HLC with the experimental data, internal validation results and statistical parameters show that the proposed QSPR model by MLR and LS-SVM methods are reliable, predictive and stable; however the non-linear model is more powerful than the linear one. (C) 2017 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.