Machine learning: Factorization Machines and Normalized Discounted Cumulative Gain for Tourism Recommender System Optimisation
Oras F. Baker, Qing Yuan · 2021
The online tourism industry has significantly developed in recent years. One prominent feature of a tourism recommender system is to assist the users in finding out potential attractions based on their preferences. A recommender system is a specific type of intelligent system, which exploits historical user ratings on items or additional information. In this research, we developed and implemented a tourism recommender system that consists of two Machine Learning algorithms. The first proposed model is the Factorisation Machines (FM), and the second model is the Normalized Discounted Cumulative Gain (NDCG) with Deep Neural Network (DNN) system, utilised to improve the search results. The evaluation shows that the FM method is effective and highly aggregated with low-frequency user rating data and could produce high-quality outputs. Finally, we expand on current trends and provide new perspectives of this new exciting development in the field of recommender systems.