Graphs for Metric Space Searching Improving Learning-object Metadata Usage during Lesson Authoring Improving Learning-object Metadata Usage during Lesson Authoring Their Great Scientific and Human Collaboration. I Am Also Grateful to My Colleagues

In order to achieve coherence and flexibility in multimedia-based learning units many authors recommend that the lesson components are structured as a graph. In a lesson graph, educational resources are encapsulated into learning objects (LO) along with their respective metadata (LOM) and are interconnected through different kinds of rhetorical and semantical relationships. Lesson graph LOs are stored within repositories where their metadata is used to ease their retrieval. Nevertheless, such systems face serious problems with LOM usage: Metadata is difficult to instantiate and lesson authors are generally reluctant to perform such tedious tasks since they do not benefit from the metadata they generate. Generating metadata automatically solves this problem. Nevertheless, this approach is limited to a restricted set of LOM attributes that generally excludes subjective metadata such as the educational characteristics of LOs. This limitation motivates this thesis to focus on a complementary method: a hybrid approach based on the synergy between automatic processes and human interventions. Hybrid LOM generation can be applied to metadata attributes that are not automatically generated. Nevertheless, it still involves the contribution of generally uncooperative users that require benefits in order to encourage their participation. We propose to study LOM usage during lesson authoring not only from the perspective of hybrid LOM generation but also from the perspective of the benefits resulting from generating LOM. Such strategy aims at supporting a positive feedback loop where the benefits motivate good-quality LOM generation while good-quality LOM increases these benefits. In particular, this thesis investigates methods for (1) seamlessly integrating hybrid LOM generation and hybrid LOM validation in a lesson authoring tool, (2) processing LOM even when some metadata attribute values are missing, incomplete, or incorrect, and (3) enhancing the performance of classical LO retrieval methods using the metadata of authored-lesson LOs. We developed an open-source lesson authoring tool, LessonMapper2, for testing the feasibility and validity of the proposals of this thesis. Preliminary experiments show that LOM generated by users significantly enhances LO retrieval during lesson authoring.