Extended Triple Store Structure Used in Recommender System

We present a database structure and main algorithms for a tourism recommender system. The presented structures are implemented in the Sights planner system, able to create plans for visiting interesting objects and events on the trip, based on the personal preference profile. The database structure can be seen as RDF extended with metainformation fields. Every fact in the database has a confidence score, interpreted as the probability. The score is used to deduct new facts and recommend objects to tourists.