Modeling the number of tourist arrivals in the United States employing deep learning networks

Abstract

Tourism plays a vital role in the U.S. economy by generating billions of dollars in revenue annually, supporting millions of jobs across various sectors such as hospitality, transportation, and entertainment. It also fosters cultural exchange and economic growth in local communities, attracting both domestic and international visitors. Considering this importance, the number of tourist arrivals in the U.S. are modeled in this study employing deep learning networks. The number of tourist arrivals in the U.S. data for the last ten years is taken from the Office of Travel and Tourism Industries and then a deep learning artificial neural network is developed in Python programming language for modeling this data. The developed deep learning network is optimized considering the accuracy of the model. Then, the deep learning network is trained using the 70% of the available data while the remaining 30% is considered as the test data. The result of the deep learning network and the actual number of tourist arrivals data are plotted together which indicate high accuracy. In order to quantify the performance of the developed model, the performance metrics namely coefficient of determination, mean absolute error, mean absolute percentage error and the root mean square error are computed. The coefficient of determination of the developed model is found to be over 0.95 indicating the high accuracy of the model. The developed model is considered as a valuable tool for the assessment of the number of tourist arrivals in the U.S. for the planning and optimizing efforts in the hospitality sector.