Recommender Systems and Social Networks: an application in Cultural Heritage

Abstract

In the last decade Recommender Systems have become useful tools helping users to find "what they need" from considerable amount of data. One of the more obvious applications of such systems in the Cultural Heritage domain is to assist users when visiting cultural environments (such as museums, archaeological sites, old town centers and so on), providing a multimedia guide that is able to dynamically suggest relevant information available in multiple web repositories (e.g. multimedia sharing systems and on-line social networks). In this paper, we propose a novel recommendation approach that combines several aspects of users -i.e. their preferences (usually in the shape of items' metadata) and interactions within a social community modeled using hypergraphs -together with items' multimedia features and context information within a general framework that can support di↵erent applications (touristic guiding services for museums, visiting paths recommendation for old town centers and archeological sites, etc.). Preliminary experiments on user satisfaction show how our approach provides very promising and interesting results.