Personalized Travel Recommendations System Using Hybrid Filtering and Deep Learning

Personalized recommendations are provided by recommender systems, which help users cope with the issue of information overload. The primary objective of this research paper is to address the issue of information overload by offering personalized tourism recommendations to users. The study proposes using multi-criteria recommendation algorithms that consider many attributes rather than depending exclusively on overall user ratings, as is common in traditional recommender systems. The fundamental concept of this strategy relies on a hybrid filtering technique. In addition, the system employs a bi-directional autoencoder (BDA) model to personalize recommendations based on various visitor profiles. The proposed methodology has evaluated various real-world data obtained from several travel websites. The experimental evaluations have demonstrated its outstanding efficacy in enhancing prediction accuracy when compared to existing algorithms. The proposed approach not only addresses the challenges associated with vast and diverse data but also offers personalized travel plans according to the unique preferences of each visitor, thus enhancing the user experience in tourist recommendation systems.