PERSONALIZING TOURIST EXPERIENCE IN SAUDI ARABIA BY LEVERAGING MACHINE LEARNING

Abstract

Saudi Arabia is rapidly emerging as a global tourist destination, driven by its vision to diversify the economy and prioritize the tourism sector. While the Kingdom offers diverse attractions, existing recommendation systems lack personalization, often providing generic itineraries that fail to cater to individual preferences. Motivated by the need to enhance the tourism experience and attract a broader audience, this study leverages machine learning algorithms to create a personalized recommendation system. The objective of this project is to address challenges in information gathering and itinerary planning by developing a hybrid recommendation model that combines collaborative filtering, knowledge-based filtering, and supervised learning. The system utilizes datasets containing user preferences, attraction profiles, and sentiment data to generate tailored recommendations. Evaluation metrics, including hit rate, precision, recall, and f1-score were used to assess model performance, demonstrating the hybrid model's accuracy and relevance compared to standalone approaches. Results indicate that the proposed system successfully delivered personalized travel plans, which simplify decision-making, and offers a unique travel experience to travelers. This study contributes to the Kingdom’s Vision 2030 by fostering innovation in the tourism sector, making Saudi Arabia a more attractive destination for global visitors. Future work will focus on integrating real-time data, multi-language support, and advanced deep learning techniques to further enhance the system's capabilities.