An adaptive conversation system to support workplace learning

This work describes the definition of a novel conversation-based workplace learning system that leverages on semantic and adaptive technologies. The proposed approach is based on the idea that conversations are commonplace human activities to learn and share knoweldge, also in the organisations, and that it is possible to exploit the organisational knowledge, represented by means of Semantic Web stack, in order to support the conversational learning processes in terms of scripting and adaptation. In particular, scripts are automatically constructed and used to guide conversation participants to the achievement of learning objectives. Moreover, two types of adaptation are provided in order to improve learning. Macro-adaptation is focused on adapting the learning experience by selecting a more suitable conversation partner basing on intermediate assessment results. Micro-adaptation is focused on adapting the learning experience by generating and providing suggestions (for conversation participants) that aim at improving the meaningfulness of learning. The main benefits of the proposed approach are the capability to improve and capitalize intentional but informal-learning experiences, to foster the organisational learning as side effect and decrease the training costs by exploiting internal skilled workers.