Hybrid Deep Learning-Based Tourism Demand Forecasting Using Big Data Analytics

Abstract

The tourism industry is increasingly relying on data-driven decision-making to gain insights into consumer behavior, manage hotel operations and enhance tourism planning. The growing volume of booking-related data offers valuable opportunities to identify demand patterns and improve prediction accuracy; however, extracting meaningful insights from complex and heterogeneous tourism data remains challenging. This study proposes an Artificial Neural Network (ANN)-based approach to predicting tourism demand using the Hotel Booking Demand Dataset. The proposed methodology analyses hotel booking information to identify significant patterns and develop an effective predictive model. Accuracy, precision, recall, and F1-score are used to assess the proposed ANN. Experimental results demonstrate that the proposed ANN achieves an accuracy of 98.91%, precision of 95.96%, recall of 99.11%, and F1-score of 98.01%. The results suggest that the ANN can effectively capture the complex relationships embedded in hotel booking data and offers reliable predictive performance. The proposed approach can therefore support accurate tourism demand forecasting and assist stakeholders in making informed, data-driven decisions for improved tourism management and planning.