Person in charge:
Lecturers and students need access to various educational materials to understand a new topic or to update their knowledge. The viewing of relationships among topics facilitates this process. However, the increasing amount of educational material available in repositories on the Internet hampers finding the appropriate content. In our research we will assist users in finding relevant courseware content in multimedia educational repositories. The project is defining and developing algorithms to find, analyze, extract hidden relationships among courseware content. The project concentrates on material presented in lectures, namely slides used during a lecture and videos of the lecture itself. These relationships will assist in the learning process and facilitate the handling of materials that are (indirectly) related to each other.
An example of analysis performed in this research. We can view the topics mentioned in courseware of different formats and the relationships between these topics.
The main goal of this research is design a methodology that combines a suite of tools to integrate courseware by highlighting the presence of relationships among the content covered by these courseware. We use graph databases as the basis of data management and to navigate among the data. The approach to classify the educational material combines ESA algorithms and the ACM Computing Classification System
Saraiva, Márcio de Carvalho; Medeiros, Claudia Bauzer
Proceedings of the 31st Brazilian Symposium on Databases, Bahia, Brazil, 2016
Though our work is general purpose, it is being tested against WebLectures, an open courseware repository at UNICAMP, with over 340 two hours lectures on Computer Science, Physics and Mathematics, and hundreds of videos and slides. Also, we will use slides and videos extracted from courses in Coursera, a web platform that provides universal access to online education material and courses from universities and organizations around the world.