Model Recommendation System for Guided Tours based on Ubiquitous and Context-Sensitive Computing

The proposed model in this research, intended to work in a guided tour context, is based on developing the tourist ontology in Python using the Owl-ready library, and describes the entities of the guided tour model. The ontology allows us to apply the concepts of ubiquity and represent context sensitivity in three ways, with geographical, temporal and environmental context. For the guided tour, the user's profile, preferences, emotional state and evaluations of the visited places are considered, as well as the profile, itinerary and site characteristics, the user's transportation preferences and the site's transportation characteristics. An ontology language was used to model the concepts and characteristics of the guided tour system, which allows inferences to be made with rules using the SWRL language with the Pellet reasoner. All models were evaluated using the RMSE metric and the accuracy, recall and F1 score metrics have been used to evaluate the predictions. This paper concludes that, among the recommender system models with collaborative filtering, the hybrid model obtained the best results for RMSE and the other metrics of accuracy, recall, and F1 score. For this reason, it is one of the most widely used recommender models in the industry.