Multi-criteria retrieval in cultural heritage recommendation systems

In recent years there has been a growing interest in mobile recommender systems for the tourism domain, because they can support users visiting new places, suggesting restaurants, hotels, attractions, or entire itineraries. The effectiveness of their suggestions mainly depends on the item retrieval process. How well is the system able to retrieve items that meet users' needs and preferences? In this paper we propose a multi-criteria collaborative approach, that offers a complete method for calculating users' similarities and rating predictions on items to be recommended. It is a purely multi-criteria approach, that uses Pearson's correlation coefficient to compute similarities among users. Experimental results evaluating the retrieval effectiveness of the proposed approach in a prototype mobile cultural heritage recommender system (that suggests visits to cultural locations in Apulia region) show a better retrieval precision than a standard collaborative approach based on the same metrics.