A comparison of two preference elicitation approaches for museum recommendations

• DOI: 10.1002/cpe.4100
• Publication Date: 2017-03-22T10:58:39Z
• Authors: Silvia Rossi; Francesco Barile; Sergio Di Martino; Davide Improta
• Keywords: Preference Elicitation; Baseline (sea)

Summary Recommendation systems based on collaborative filtering methods can be exploited in the context of providing personalized artworks tours within a museum. However, to effectively used, we have several problems addressed: user preferences are not expressed as rating and recommendation must provide for new users efficient simple elicitation processes that do require much effort time. In this work, present evaluate 2 state‐of‐the‐art approaches share aim rely individual item ratings. The first method uses clustering algorithm categorize items recommendations, while second one is inspired by matrix factorization approach select couples groups obtain preference profiles. We with both an off‐line simulation study find optimal configuration well effectiveness proposed methods. Results show permit profiles time substantially less than baseline method, differences terms prediction accuracy minimal.