【Objective】Based on a multimodal deep learning method, this paper analyzes consumer sentiment through travel reviews, which predicts hotel stock movement of hotel

【Methods】In this paper, a multi-modal deep learning model is constructed. First, the multi-modal information is encoded. Furthermore, the interaction information between text and images is extracted through LSTM and graph neural network. Finally, the hotel stock prediction is performed.

【Results】Empirical research was carried out based on travel review data from Yelp, and compared with related baseline models. The experimental results show that the multimodal model proposed in this paper has advantages, and the average accuracy of stock prediction achieves 59.10%.

【Limitations】The proposed model is only tested on the dataset of four hotels on the Yelp website, this paper has not been further validated on other travel platforms.

【Conclusion】The proposed model can fully extract the interactive information between different modalities, which effectively improves the accuracy of hotel’s stock prediction.