In order to find the most appropriate combination of base classifiers for the classification task, an AHP-based base classifier selection method for Stacking algorithm is proposed. The weights of the single performance index of the classifier are determined by the AHP, and the base classifier is selected by taking the weighted sum result of each index as the standard for evaluating the quality of the classifier. Through the Stacking integration of different combinations of base classifiers, it is found that the single classifier with the best performance is not the best, and the model with the largest number of base classifiers is not the most suitable for the task requirements. The results show that the method is feasible and effective in improving the performance of the classifier.