Course and exercise sequencing using metadata in adaptive hypermedia learning systems
Journal on Educational Resources in Computing (JERIC)
Modern Information Retrieval
A network-based approach to text handling for the on-line scientific community
A network-based approach to text handling for the on-line scientific community
An efficient boosting algorithm for combining preferences
The Journal of Machine Learning Research
Automatic Generation of Metadata for Learning Objects
ICALT '06 Proceedings of the Sixth IEEE International Conference on Advanced Learning Technologies
Hybrid System for Generating Learning Object Metadata
ICALT '06 Proceedings of the Sixth IEEE International Conference on Advanced Learning Technologies
Taking advantage of the Semantics of a Lesson Graph based on Learning Objects
Proceedings of the 2007 conference on Artificial Intelligence in Education: Building Technology Rich Learning Contexts That Work
Hi-index | 0.00 |
Many authors believe that in order to achieve coherence and flexibility at the same time in multimedia-based learning units, it is highly recommendable to structure the different components as a graph. In a lesson graph, educational resources are encapsulated into learning objects (LO) along with their respective metadata and are interconnected through different kind of rhetorical and semantical relationships. The LOs of these graphs are stored within repositories, where their metadata are used to ease their retrieval. In this paper we propose to integrate the processes of searching LOs and editing the lesson graph. This new framework extends traditional keyword and metadata search to take advantage of the information stored implicitly in the lesson graph structure, making LOs retrieval more effective and the expression of queries more intuitive. The retrieval of the learning material consists of two processes: (1) The user first defines the topological location of a required piece of educational material within the lesson graph, this is, its relationships with other pieces. (2) Then, the user issues a traditional keyword query, which is processed by an IR system modified to take the graph structure into account. Experiments show the advantages of this approach.