MF-for-UPSuM 2012 - Workshop on Matrix Factorization techniques for student skills and user preference modeling

Workshop on Matrix Factorization Techniques for User Preference and Skills Modeling
The First Workshop on Matrix Factorization for User Preference and Skill Modeling (MF-for-UPSuM workshop) aims at bringing together researchers from the recommender and student modeling fields who are working on Matrix Factorization (MF) techniques. These techniques have gained a strong acceptance in the recommender systems field in the last few years, in good part due to their strong success and the visibility gained through the Netflix contest in 2009. They have recently been applied in the field of student skills modeling, also gaining strong success by matching the state of the art performance of well established approaches to predict student performance.
The aim of this event is to offer an opportunity to specifically focus on MF techniques and build bridges with the researchers working on user preference modeling from the recommender system community, and the researchers from the student skills modeling community. These two communities are central to the UMAP conference.
Submissions of original on-going works and previously unpublished research related to the Workshop topics are requested. We also welcome position and review papers that can have interest to an audience who want to learn more from the applications of the techniques within the two target fields.
The Workshop encompasses the following topics:
Linear models of skills and preferences
Factorization models
Temporal dynamics
Learning algorithms
Factors selection
Validation issues and experimental comparisons
Assessment of uncertainty
Ensemble models
Applications