Recommender systems make learning more fun

Recommender systems for learners in informal learning networks make it possible to choose a personalized learning path and can make learning more interesting. Says Hendrik Drachsler, researcher at CELSTEC in his PhD thesis. He'll defend the thesis on 16 October 2009 at the Open Universiteit Nederland in Heerlen.

Informal learning
If you want to develop yourself, there are several ways to gain knowledge. You can choose a regular study programme or opt for distance education. Relatively new is learning through informal networks on the internet. You form part of a community. There are no teachers or books that guide you through the content within a certain space of time. You choose your own pace, set your own goals, pick your own learning activities. The learners inform and advise each other. The problem of informal learning is that the internet offers an enormous quantity of information. It’s not easy to search and find information and learning activities that match with your individual knowledge and preferences.

Combination of techniques
For the EU project TENCompetence Hendrik Drachsler searched for the most suitable way to recommend learning activities, taking into account the personal needs and preferences and also the learning goals of the members of the community. He looked into commercial navigation and recommendation systems for e-commerce activities, like the systems of Amazon.com. And he looked into systems for regular students. Drachsler concluded that the specific behavior of community members asks for a combination of different recommendation techniques. On the basis of the most appropriate techniques, he developed a prototype navigation system. He set up experiments to determine whether such a system does de facto offers an added value to the target group.

More satisfaction, less study time
The conclusion was that the system had a significantly positive influence on the study time. The experiment group needed less time to complete a same number of learning activities. Furthermore the participants of the experimental group chose more personalized learning paths than members of the control Group, because they explored more learning paths. A second experiment did not show differences between the performance of the experimental and the control groups, but the satisfaction in the experimental Group was clearly higher and the they needed less study time to reach their learning goals.

Best paper
 The results led to the development of a second prototype, ReMashed. This prototype shows how recommendation systems can benefit from the information generated in web 2.0 by life-long learners in a learning network. It recommends the most suitable sources, based on tags and evaluations of individual learners. Furthermore the learners can specify learning goals and prior knowledge by self assessment. For his description of this prototype in a paper Drachsler received a best paper award from the European project Role, that is specifically aimed at the development of open learning environments.

Hendrik Drachsler defends his PhD titled Navigation Support for Learners in Informal Learning Networks on Friday 16th October 2009 at 13.30 hours at the de Open Universiteit Nederland in Heerlen. Promotor is prof. dr. E.J.R. Koper, co-promotor dr. H.G.K. Hummel.