Combining semantic and collaborative recommendations to generate personalized museum tours

Abstract

Our work takes place in the field of support systems to museum visits and access to cultural heritage. Visitors of museums are often overwhelmed by the information available in the space they are exploring. Therefore, finding relevant artworks to see in a limited amount of time is a difficult task. Our goal is to design a recommender system for mobile devices that adapts to the users preferences and is sensitive to their contexts (location, time, expertise…). This system aims to improve the visitors’ experience and help them build their tours onsite according to their preferences and constraints. In this paper we describe our recommendation framework, which consists in a hybrid recommendation system. It combines a semantic approach for the representation of museum knowledge using ontologies and thesauruses with a semantically-enhanced collaborative filtering method. A contextual postfiltering enables the generation of a highly personalized tour based on the physical environment, the location of the visitors and the time they want to spend in the museum. This work is applied to the Compiègne Imperial Palace museum in Picardy.