A Framework for the Domain-Independent Collection of Attention Metadata Maren Scheffel, Martin Friedrich, Katja Niemann, Uwe Kirschenmann, and Martin Wolpers Fraunhofer Institute for Applied Information Technology, Schloss Birlinghoven, 53754 Sankt Augustin, Germany {maren.scheffel,martin.friedrich,katja.niemann, uwe.kirschenmann,martin.wolpers}@fit.fraunhofer.de http://www.fit.fraunhofer.de Abstract. We present a simple and extendible framework to collect at- tention metadata and store them for further analysis. Currently, several metadata collectors have been implemented but the framework allows the easy integration of further data collectors. Analysis results, e.g. rec- ommendations or fostering of self-reflection, can then support the user in her learning. Keywords: attention metadata, framework architecture, recommendation. 1 Introduction Today, learners spend significant amounts of time on continuously managing digital information that is either found or provided. In consequence, learning is severely hampered through the continuous distractions of provided content that simply increases the cognitive load of learners beyond meaningful states. The information provision is basically controlled by the information coming into the cognitive system. The critical element is the attention that enables the utilization of this type of processes. Making the important information more salient thus directs cognitive resources to the most valuable and critical learning material. Furthermore, by capturing indicators for attention, supportive learning tools become possible, e.g. by enabling support for reflective learning [1]. Recent learning environments are highly specialized to support the user in specific tasks in specific environments. Task and environment independence are not addressed with these applications. Hence, the applications only insufficiently tailor their support to the needs of the learner. Furthermore, through relying on information about the user collected from the system internally, all other information about the learner is neglected. In consequence, when the learner changes the learning environment, only such information about the learner is transferred that is created using user profiling standards like PAPI [2]. These standards do not provide means to describe all aspects of the user. Therefore, we have taken a different approach by collecting contextualized attention metadata M. Wolpers et al. (Eds.): EC-TEL 2010, LNCS 6383, pp. 426–431, 2010. c  Springer-Verlag Berlin Heidelberg 2010