10 Mar 2021
14:00 Master's Defense Fully distance
Theme
Semantic-enhanced recommendation of video lectures relying on ontology-based annotations
Student
Marcos Vinícius Macêdo Borges
Advisor / Teacher
Julio Cesar dos Reis
Brief summary
Learning support systems exploit diverse multimedia resources to consider student individualities as well as different learning styles. However, the growing amount of educational content available in different formats and in a fragmented way makes it difficult to access and understand the concepts under study. Although the literature has proposed approaches to explore recommendation techniques that allow explicit representation of semantics through artifacts such as ontologies, this line has not been fully explored and still requires many research efforts. This research aims to conceive a method of recommending educational content exploring the use of semantic annotations on textual transcriptions of video lessons. Annotations serve as metadata that express the meaning of excerpts from the lessons. The recommendation technique, as the main expected contribution, is based on the available notes to define strategies for ranking available content from the semantic proximity of the concepts combined with machine learning techniques. The contribution involves the development of functional software prototypes for experimental validation based on real video class content and should highlight the main advantages and limitations of the approach. The results obtained will allow access to the most appropriate recommendations to improve the learning process, presenting the possibility of a more satisfactory experience by students.
Examination Board
Headlines:
| Julio Cesar dos Reis | IC / UNICAMP |
| Flavia Linhalis Arantes | NIED / UNICAMP |
| Diego Addan Gonçalves | IC / UNICAMP |
Substitutes:
| Juliana Freitag Borin | IC / UNICAMP |
| Ricardo Edgard Caceffo | UNIVESP |