Restructuring Ontologies through Knowledge Discovery

In this work, we describe an approach for maintaining self-adapting personalized ontologies. We motivate this work with the observation that users are often interested in only small parts of larger domain ontologies; however, we also observe that users are frequently unaware of concepts that may be of interest given their past preferences. As such, we suggest that personalized ontologies should be continuously specialized and generalized to represent only those concepts that have proven to be of value to the user.