A data-driven framework for tourism resource allocation using social spider optimization and fuzzy neural networks

Abstract

The increasing use of data-driven technologies has significantly influenced how tourism systems are managed, particularly in the context of resource allocation and demand prediction. However, tourism environments remain highly dynamic, where factors such as seasonal variations, event-driven demand, and environmental conditions introduce uncertainty and complexity. Addressing these challenges requires models that can both capture nonlinear relationships and adapt to changing conditions. In this study, a hybrid approach named Social Spider Optimization-based Tourism Resource Allocation with Fuzzy Neural Network (SSO-TRFNN) is proposed. The model combines the optimization capability of Social Spider Optimization with the uncertainty-handling nature of fuzzy neural networks, enabling more effective modeling of tourism data. An exploratory analysis is first conducted to examine the influence of key variables, including visitor count, flight arrivals, temperature, and major events. The proposed model is evaluated against several existing approaches such as Genetic Algorithm, Particle Swarm Optimization with Neural Network, Artificial Neural Network, Fuzzy Logic system, and LSTM-based deep learning. The results show that the proposed model achieves an accuracy of 96.1%, precision of 95.4%, recall of 94.8%, and an F1-score of 0.95. It also demonstrates higher resource allocation efficiency (92.7%) while maintaining a lower computational time of 1.95 s. Overall, the findings suggest that the proposed SSO-TRFNN model provides a practical and effective solution for tourism resource optimization, with the potential to support smarter and more responsive tourism management systems.