“…ML models have the ability to learn complex nonlinear data patterns, and they can be effectively used for the estimation or prediction of state variables using easily available descriptor data . Although ML models have gained a lot of attention in recent years in the fields of finance, telecommunications, language processing, etc., the studies reporting the ML models for the prediction of CO 2 solubility in different solvents are very few. − Further, ML models have been employed for carbon capture, utilization and storage, discovery of porous materials for carbon capture, prediction of CO 2 absorption, prediction of thermodynamic properties of CO 2 in the solvent, study of the correlations between physical properties (density, viscosity, and specific heat capacity) and CO 2 loading, prediction of mass transfer coefficients in the CO 2 absorber, − design of new solvents, , etc. Furthermore, there are a few studies reporting ML models for the prediction of CO 2 solubility in different, physical, chemical, and ionic solvents, and the corresponding summary points are presented in Table , including the details on the type of CO 2 solvent, the ML model developed, input data used for model building, etc.…”