In response to the problem of poor accuracy in predicting daily tourist flow in tourist attractions, this study proposes a daily tourist flow prediction model based on stacking ensemble learning algorithm. Firstly, preprocess, select features, and deduplicate the historical tourist flow data collected from the scenic area. Then, using the Stacking ensemble learning algorithm, the prediction results of multiple base learners such as random forest, gradient boosting decision tree, extreme gradient boosting, K-nearest neighbor, etc. are used as new feature inputs to the meta model. Finally, in the meta model, logistic regression algorithm is used to train and output the final prediction results. The experiment shows that the model has high prediction accuracy and efficiency, and the Stacking algorithm performs well, providing strong support for scenic area planning and management.