Present position: Temporary Researcher
|Thesis title:||Methodologies and Techniques for Cooperative Learning Environments with Users Support and Evaluation|
|Research area:||Advanced Software Architectures and Methodologies|
This work describes Virtual Campus, an eLearning platform developed at Politecnico di Milano and designed to cope with these weaknesses.
Virtual Campus relies on and extends the SCORM standard. We see SCORM as a good opportunity to support interoperability among eLearning tools since it enables the definition of a data model that can be shared among them. However, we have noticed some weaknesses in such a data model. These weaknesses mainly concern the way instructional materials can be described by metadata, structured and made available for reuse.
In our vision, composition is a key-concept: All the instructional resources should be described by proper metadata and recursively composed. Thanks to the recursive composition mechanisms, reuse both within a single platform and among platforms can be greatly enhanced. Moreover, the definition of proper metadata can support not only browsing and reuse of materials, but also installation and use of them.
The Virtual Campus project aims at providing an implementation of the aforementioned concepts, providing an enhanced set of metadata, as well, as a novel authoring methodology. In particular, the authoring phase is accomplished at three levels of abstraction. Starting from an abstract logical representation, a course undergoes a customization process whose goal is to best fit the teachers' specific requirements, while increasingly transforming the course representation into one that can be executed by means of a workflow management system.
The workflow engine enhances the SCORM run-time environment, guiding Learners through the instructional paths.
In the context of Virtual Campus, we also developed a tutoring and validation module. A profiling tool instruments the runtime engine, tracking the learners' behavior within the Virtual Campus environment and generating a probabilistic user model. The model is defined in terms of learning attitudes, efficiency of learning strategies, and attitude to cooperation and communication.
Derived information is used to generate and make available graphical reports to teachers. Moreover, an automatic tutor exploits the user model to support learners' choices, making suggestions which take into account the style, the behavior and the results obtained in the past.
Users may exploit several different eLearning systems. In that case, a user model in a given platform is a partial fragment of an overall virtual user model. The fragments have then to be collected and merged into a global user profile. We investigated algorithms able to cope with distributed, fragmented user models.