Estimation and prediction of the air-water interfacial tension in conventional and peptide surface-active agents by Random Forest Regression

•A Random Forest model evaluated via a fivefold cross-validation was put forward.•The model was used for estimating the air–water interfacial tension of various surface-active species.•A database with the air–water interfacial tension of 691 surfactants was used for training the model.•Structure-performance relationships were obtained for surface active molecules.•Sequences of amino acids are proposed for screening peptides with high interfacial activity.