|
چکیده
|
In this work, quantitative structure–retention relationship (QSRR) approaches were applied for modeling and prediction of the gas chromatographic retention indices of some amino acids (AAs) and carboxylic acids (CAs). The genetic algorithm (GA) method was used to select the most relevant descriptors, which are responsible for the retention of these compounds. Then, multiple linear regression (MLR), artificial neural network (ANN) and support vector machine (SVM) were utilized to construct the nonlinear and linear quantitative structure–retention relationship models. The obtained results revealed that the GA-ANN developed model was better than other models. This model has the average absolute relative errors of 0.043, 0.052 and 0.045 for training, internal and external test set. Applying the 10-fold cross-validation procedure on GAAAN model obtained the statistics of Q2 = 0.941 which revealed the reliability of this model.
|