Authors

Joris Klerxk
Hannes Ebner

Title

Contextual Open Educational Resources for Future Recommender Scenarios

Published in

dataTEL workshop – Data Sets for Technology Enhanced Learning. Alpine Rendez-Vous. La Clusaz, France, 29-31 March 2011.

Abstract

ARIADNE is a European foundation that aims to foster “Share and Reuse” of learning resources. Reusing digital resources for learning has been a goal for several decades, driven by potential time savings and quality enhancements. To support this goal, ARIADNE has created a standards-based infrastructure for managing learning objects in an open and scalable way. In the last decade, we have collected a large dataset of more than a million educational learning resources from numerous content providers all over the world. It is now essential to work on novel and flexible ways to interact with this data, using contextual information such as location, proximity, task, mood, etc. End users should be able to access this rich information space in various ways such as

• Recommender systems and information filtering,

• Visualization of concrete and conceptual information,

• Mobile context aware computing,

• Multi-touch navigation on large displays, and

• Mash-up applications that blend together this dataset with existing information in other information spaces, resulting in Personal Learning Environments or similar.

This paper will specifically focus on the first part and will show how recommender and information filtering systems can take advantage of this large data set of educational learning resources to provide a modern personalised learning landscape. This paper will start with a description of the dataset. Secondly, we will present how we make this dataset available to all actors in TEL in 2 different bindings:

• Learning Object Metadata (IEEE LTSC LOM), and

• LOM instance expressions using the Dublin Core Abstract Model (DCAM).

The conceptual and technical description of the available interfaces that can be used to access the dataset will be presented in detail. This paper concludes with a discussion on the use of this dataset for recommender systems in future TEL scenarios.