Interoperability between AEH user models

Creating a good quality Adaptive Educational Hypermedia (AEH) system involves a great expenditure of time, effort and money, especially to author the materials in order to implement adaptivity. Moreover, authoring AEH systems is as hard as maintaining them. Furthermore, with the new advances in AEH development, the authored content in a given system can become outdated very rapidly. When the massive overhead in authoring is combined with the possibility of the current AEH system becoming unavailable due to lack of maintenance, providing interoperability of adaptive systems becomes necessary. This paper describes our research on providing interoperability of user models between different AEH systems via a one-to-one conversion between two AEH systems, MOT and WHURLE. Here, we identify the differences and commonalities and address how these influence the efficiency of the conversion of the two systems' user models. Although this conversion is done via simple peer-to-peer interaction, it can be easily extended to make use of semantic web technologies, as for example RDF or XML conversion formats, or web services for user model exchange. Therefore, we consider that our work on feature extractions sets the basis for such conversions for the Web and Semantic Web.