A two-sided travel marketplace is an E-Commerce platform where users can both host tours or activities and book them as a guest. A travel recommender system needs to both understand characteristics of its inventories, and to know the preferences of each individual guest. In this work, we present our efforts on building a recommender system for Airbnb Experiences, a two-sided online marketplace for tours and activities.
• Knowledge Graph Expansion for Tourism Applications: Classic knowledge graph research focuses on building concepts that are more generic, and domain-specific knowledge graph research rarely discusses topics related to travel and tours. By contrast, we precisely focus on the travel domain that involves tightly with locations, and extend the generic terms into more location-specific concepts.
• Recommendation with Limited Data Availability: An additional information we can utilize is user profiles, such as travel destination and user origin. Due to lack of user-item interaction data, we find directly using the categorical information easily leads to overfitting. Instead, we propose a novel method of dealing with categorical features.