LAUSR

Abstract In this communication, a mathematical approach on the basis of Leverage algorithm has been applied for evaluating the quality of experimental data of CO 2 loading capacity of DEA aqueous solutions as well as detecting the probable doubtful data points. Furthermore, the proposed scheme determines the applicability range of the employed model. First, capability of several techniques including MLP-ANN, GA-RBF, CSA-LSSVM, Hybrid-ANFIS, PSO-ANFIS, and CMIS in representing the equilibrium absorption of CO 2 in DEA solution have been evaluated. Results of modeling study reveal that all the developed models are accurate and reliable for the application of interest. However, the presented CMIS model outperforms other models thanks to higher value of correlation coefficient and more accurate predictions. Finally, it was found that the developed CMIS model is statistically valid, and all the data are within the range of applicability of the model. Further to the above, only one data found to be probable doubtful.