Fonte: Turismo y Sociedad; Vol. 34 (2024): Enero-Junio; 149-178
Palavras-chave:
Resumo: The level of satisfaction of a tourist with the destination visited, as well as his or her intention to revisit the destination, is assumed to be dependent on his or her previous experience with the place. To observe this relational perspective, a dataset of 386 tourists who visited the city of Medellin (Colombia) in 2018 was used. To predict the variables of revisiting the city and satisfaction with the destination, we consider push and pull variables. Four statistical learning models were estimated to classify tourists: Logistic Regression (LR), Random Forests (RF), Support Vector Machines (SVM), and the Extreme Gradient Boosting algorithm. The most important variables in the satisfaction estimation were: ‘talk about future travel experiences’ and ‘go to places my friends have not visited’, while for revisiting the city the variables were: ‘visit historical places’ and ‘travel at low prices’.
