May 28, 2010 by Hendrik Drachsler
Comments (0)
*************************************************************
FIRST CALL FOR TEL DATASETS
dataTEL Cafe event for sharing datasets for Technology-Enhanced Learning
Barcelona, Spain, 29 September 2010
Organised jointly by the Workshop on Recommender Systems for Technology-Enhanced Learning (RecSysTEL) and the dataTEL Theme Team.
at the 4th ACM Conference on Recommender Systems (RecSys 2010)
and the 5th European Conference on Technology Enhanced Learning (EC-TEL 2010)
Detailed information at:
http:/
http:/
*************************************************************
AIM & TOPICS
Recommender systems are increasingly applied in TEL research to personalise learning content and to find suitable peer learners according to their individual needs, preferences and learning goals. In the world of commercial recommender systems, it is a common practice to use public available datasets with specific characteristics (e.g. MovieLens, Book-Crossing, or the EachMovie dataset) to evaluate recommender system algorithms. These datasets are used as benchmarks to develop new recommendation algorithms and compare their effects to other algorithms in a standardised way. The representation of implicit or explicit feedback from users is required to produce suitable recommendations. This feedback can be in several forms. For example, in the case of collaborative filtering systems it can be ratings or votes (i.e. if an item has been viewed or bookmarked). In the case of content-based recommenders, it can be textual reviews or simple tags (keywords) that users provide for items. In addition, information is rehired to identify who provided feedback (user ID) and upon which item (item ID). The user-rating matrix used in collaborative filtering systems is a well-known example.
Although there is a lot of research conducted on recommender systems in TEL, they lack standardised ways to evaluate the performance of different recommendation systems using standardised (interoperable and reusable) datasets. At the momenten, there is not one dataset publicly available neither for formal learning nor for non-formal learning settings. Thus, it is not possible yet for TEL recommender systems' researchers to apply and benchmark their algorithms on existing, public datasets.
The dataTEL Theme Team addresses these challenges by discussing the topic on various scientific events. We aim to launch the first TEL dataset competition in 2011 where researchers can compete in developing the first and most effective TEL recommender systems for available TEL datasets. For this aim we invite relevant groups to submit their existing datasets from TEL applications (e.g. datasets of ratings, tags, reviews, etc.) that can be used as input for TEL recommender systems.
The most suitable datasets will be announced at the dataTEL Cafe event at the EC-TEL conference 2010 in Barcelona. The winner of the dataTEL challenge will receive a best TEL dataset award!
IMPORTANT DATES
SUBMISSIONS
* two paper description: 2 pages and dataset file (specs/format to be announced soon)
Submission will be available through the EasyChair submission system:
http:/
If you haven't an EasyChair account yet, you'll be asked to create it before you can access the RecSysTEL'10 page.
STEERING COMMITTEE
to be announced soon.
CO-CHAIRS
* Hendrik Drachsler, Open Universiteit Nederlands, The Netherlands
* Nikos Manouselis, Greek Research & Technology Network, Greece
* Riina Vuorikari, European Schoolnet, Belgium
* Katrien Verbert, Katholieke Universiteit Leuven, Belgium
* Stephanie Lindstaedt, Know Center, Austria
* Martin Wolpers, FIT Fraunhofer, Germany
Feel free to spread the word using the 'dataTEL' tag!
ABOUT RecSys 2010 AND EC-TEL 2010
The 4th ACM Conference on Recommender Systems (RecSys 2010) is the premier annual event on research and applications of recommender technologies. It will promote a close interaction among practitioners and researchers, reaching a wider range of participants including those from Europe and Asia. See http:/
The 5th European Conference on Technology Enhanced Learning (EC-TEL 2010) brings together technological developments, learning models, and implementations of new and innovative approaches to training and education. The conference traditionally explores how the synergy of multiple disciplines can provide new, more effective and more especially more sustainable, technology-enhanced learning solutions to learning problems. See http:/
*************************************************************
