Meta-rules: Improving adaptation in recommendation systems

Abstract

Recommendation Systems are central in current applications to help the user find useful informat ion spread in large amounts of post. videos or social networks. Most Reconunendation Systems are more effective when huge amounts of user data are available in order to calculate similarit ies between users. Educational applications are not popular enough in order to generate large amount of data. In this context, nile-based Reco tumendarion Systems are a better solution. Rules are in most cases written a priori by dom ain experts: they can offer good recomiuendat ions with even no application of usage informat ion. However large rule-sets are hard to maint ain. reengineer and adapt to user goals and pref erences. Meta-rules. rules that generate rules, can generalize a rule-set providing bases for adaptation. reengineering and on the fly generat ion. In this paper. the authors expose the benef its of meta-rules implemented as part of a metar ule based Recommendation System. This is an effective solution to provide a personalized reco mmendation to the learner, and constitutes a new approach in rule-based Recommendation Systems.