Demand forecasting is important for management and decision making in the tourism sector. However, research on deep learning forecasting, which combines multiple data sources, is still in the development phase. This study proposes a bidirectional long short-term memory (BiLSTM) forecasting method incorporating an attention mechanism (ATT-BiLSTM) that can better extract data features from a set of predictor variables consisting of multiple predictor variables (generated from historical tourist volume, search engine data, weather data and day off data). The research experimentally validates the effectiveness of the method using the famous Chinese tourist attraction Jiuzhaigou as a case study. The results show that the proposed model not only has better generalization ability but also significantly outperforms the four benchmark models, convolutional neural network (CNN), SVR, LSTM, and BiLSTM, in terms of prediction accuracy. In addition, we analyse the importance of the different predictor variables in the prediction model characteristics.

Few studies have investigated tourism demand forecasting from a mobility perspective. We explore whether population mobility data improve the ability to forecast tourist volumes in scenic spots. We selected Jiuzhaigou and Mount Siguniang in China as study cases and ensured the robustness of the results using single- and multistep rolling evaluation techniques. The findings reveal a significant correlation, long-term cointegration, and two-way Granger causality between intercity population flow and tourist flow. The comparison model with population flow as a variable outperforms the benchmark model. We prove the feasibility of leveraging population flow in tourism demand forecasting and suggest further research.