Collaborative Filtering Enhanced by Demographic Information for Tourist Sites Recommendations

Abstract

Recommender systems are very applicable in the tourism domain to alleviate the information overload problem. These systems are being developed to help tourists find sites and services that are of interest to them. To promote tourism in a city, this paper presents an enhanced user-based collaborative filtering approach with demographic information for recommending touristic sites, which provides precise recommendations. The collaborative filtering approach recommends touristic sites considering two user types: new tourist and registered tourist. In this way, the recommendations of tourist sites are generated according to the user's historical rating or based on the user's demographic information. The approach is evaluated on a tourist sites dataset extracted from the TripAdvisor platform that contains historical ratings, and demographic information of the tourist. To measure the performance of the proposed approach, RMSE is used to evaluate predictions accuracy, and Precision and Recall measures to evaluate the quality of recommendations. The results showed an improvement in the prediction accuracy and a significant performance with precision and recall, especially in the presence of the cold-start problem.