Global warming is a major concern in environmental management. Among the greenhouse gases, carbon dioxide plays an important role in global warming. Many novel technologies offer different solutions to reduce carbon dioxide emissions to the environment. One of the most commonly used technologies to absorb carbon dioxide is the use of solvents, especially amine solutions. The chemical structure of amines has a powerful effect on carbon dioxide absorption. Due to the fact that laboratory studies on a large number of amines are costly and time-consuming, the modeling of the effect of different structural parameters on the absorption capacity of amine solutions plays an important role. Quantitative structure property relationship (QSPR) provides an effective method for predicting amines capacity for carbon dioxide absorption. In this work, the density functional theory (DFT-B3LYP) method was employed for optimization of the molecular geometries. GA-MLR (genetic algorithm-multilinear regression) and LS-SVM (least-squares support vector machines) were utilized to build linear and nonlinear QSPR models, respectively. The value of the square of correlation coefficient (R-2) for the MLR and SVM models are 0.962 and 0.992, respectively. The suggested models were evaluated by a variety of statistical tests. The results showed that using a nonlinear model is better than a linear model for predicting the amine capacity to absorb carbon dioxide. Based on the structural information that were appeared in the model,we concluded that the volume of the amine molecules and the length of the carbon chain in the structure of amines have significant effects on amine capacity. Comparing this work with others showed that this model is very accurate and versatile to the prediction of carbon dioxide absorption.