This study aims to implement a tourism information management system based on the concept of Smart Tourism. As the people's living conditions have improved, the traditional model of tourism management failed to match the customers' demands, because it has some flaws such as low efficiency. Integrating AI with tourism management helps to meet clients' needs with smarter and more comfortable accommodations before, during, and after their trip. This study optimizes the artificial neural network (ANN) by utilizing long short-term memory (LSTM) to address its inherent faults, such as delayed convergence, easy oscillation, and falling into local minima. The experimental findings demonstrate that this algorithm achieves a recall rate of 99.6% and a prediction accuracy of 98.2%. This algorithm offers some useful recommendations for tourist management while avoiding the pitfalls of more conventional algorithms. In addition, this study extends the idea of "smart tourism" to particular forms of tourist informatization, which has the potential to boost the tourism industry's overall development level by facilitating structural upgrades and transformations.
Read full abstract