Combining Situation and Content Similarities in Fuzzy Based Interest Matchmaking Mechanism
Younès Djaghloul, Valentin Grouès, Yannick Naudet · 2012
In this paper, we present an approach to weight the influence of context regarding user interests in content-based recommendations. The balance between interest-content matching and situation matching is formalized using a Fuzzy model, allowing more intuitive personalization. Using content consumption history, this model can be optimized from an initial general model to a new one fitting better the preferences of the user. We propose a generic service for such personalized context-aware recommendations. Experimental results are discussed and show the effectiveness comparing to a classical weighted function.