Getting to know your user–unobtrusive user model maintenance within work-integrated learning environments

Work-integrated learning (WIL) poses unique challenges for user model design: on the one hand users’ knowledge levels need to be determined based on their work activities – testing is not a viable option; on the other hand users do interact with a multitude of different work applications – there is no central learning system. This contribution introduces a user model and corresponding services (based on SOA) geared to enable unobtrusive adaptability within WIL environments. Our hybrid user model services interpret usage data in the context of enterprise models (semantic approaches) and utilize heuristics (scruffy approaches) in order to determine knowledge levels, identify subject matter experts, etc. We give an overview of different types of user model services (logging, production, inference, control), provide a reference implementation within the APOSDLE project, and discuss early evaluation results.