Navigating through the complex transport system: A heuristic approach for city tourism recommendation

The fast development of machine learning and artificial intelligence has led to a great improvement of the smart tourism recommendation system, however many problems associated with the choice of transport modes in city tourism have yet to be solved. This research attempts to address this issue by proposing a model of customized day itineraries with consideration of transport mode choice. With improved particle swarm optimization and differential evolution algorithm, a nondominated sorting heuristic approach was devised. A case study was carried out in Chengdu, China to examine the performance of our approach. The results show that compared with extant methods, our approach achieves better performance. In addition, our approach can create more sensible, multifarious, and customized itineraries than previous methods. Tourism organizations and mobile map app providers could integrate our proposed model into their existing smart service systems, as part of their e-business or digital strategy for enhancing tourist experience.