Ontology-based rules for recommender systems

Jeremy Debattista, Simon Scerri, Ismael Rivera, Siegfried Handschuh

2012 · 14 citations · 8 references

Concepts

Abstract

Abstract. Nowadays, smart devices perceive a large amount of information from device sensors, usage, and other sources which contribute to defining the user’s context and situations. The main problem is that although the data is available, it is not processed to help the user deal with this information easily. Our approach is based on the assumption that, given that this information can be unified in a single personal data space, it can be used to discover and learn rules to provide the user with personal recommendations. In this paper we introduce a Rule Management Ontology to support the representation of event-based rules that trigger specific actions. We also discuss how a context listener component can provide recommendations based on the perceived context-data, or in the future, semiautomatically learnt rules. 1

References

8