A Personalized Hybrid Tourism Recommender System

This paper focuses on building personalized recommender system in the tourism field. The application recommends to a tourist the best attractions in a particular place according to his preferences, his profile and his appreciation to previous visited places. This paper proposes a hybrid recommender system that combines the three most known recommender methods which are: the collaborative filtering (CF), the content-based filtering (CB) and the demographic filtering (DF). In order to implement these recommender methods, we have applied different machine learning algorithms which are the K-nearest neighbors (K-NN) for both CB and CF and the decision tree for the DF. The hybridization is a good choice to make the best of their advantages and to overcome the cold start problem. To enhance the recommendation accuracy, we use two hybridization techniques: switching and weighted. For the weighted approach, a novel linear programming model is applied to obtain the optimal weights' values. An extensive experimental study is conducted based on different evaluation metrics using extracted data from TripAdvisor. Our results show that the hybrid method is more accurate than the other recommender approaches used separately.