Since the selectivity of an ion selective sensor is directly related to the stability constants of ion–ionophore complexes, we predicted the complexation stability (K) of cerium ions with different ionophores by the quantitative structure–property relationship model. Genetic algorithm (GA) feature selection approach was selected to choose the proper molecular descriptors which were then subjected to multiple linear regression (MLR) for prediction of the log K. The predictive ability of the built genetic algorithm-multiple linear regression (GA-MLR) model was evaluated using Leave-one-out cross-validation, Leave-group-out cross-validation, Y-randomization, and test set compounds. Statistical parameters of the model (R²_train=0.852, Q²_LOO= 0.813, and Q²_LGO=0.777) indicated the ability of the GA-MLR model to predict the response of ionophores in cerium-selective sensors based on complex stability constants. Also, the applicability domain of the model was analyzed by the Williams plot. Based on this study, some key features are identifiable to appraise the selectivity of cerium sensors that can be used to design new selectophores.