Home »

Sujet de thèse - 2022

PhD Integrating and exploring linked educational resources

Niveau : Doctorat

Période : 2022-2025

Context and motivation                               (more details here)

Teachers have been digitizing their courses for a while and the ongoing digital transformation was accelerated by the Covid-19 lock-downs. Teachers usually search for open educational resources (OER) on the Web to reuse and combine in a course. There are many available, useful, and pertinent resources (slides, videos, figures, text, code, etc.), but finding them and organizing them in a course plan is challenging. Ideally, the necessary analysis of available resources to match a course plan and the licenses verification should not be time-consuming.

Thanks to semantic web technologies, this work aims to allow teachers to define a sketch of a new course from which a set of relevant and license compatible educational resources will be suggested for her course. The course sketch may contain metadata such as the intended license of the course, learning outcomes, the knowledge required, knowledge attempted, skills expected, an initial course syllabus, expected duration, targeted competencies, etc. Machine-readable semantic annotations will help link and enrich educational resources thanks to well-known ontologies. 

Problem statement

A compatibility graph of licenses [1] can allow producers of educational resources to know which license(s) can protect a combination of resources. When licenses of combined resources are incompatible, it is not possible to license the course. In that case, it is necessary to discard resources that are protected by conflicting licenses. However, this may lead to a query with empty results, i.e., the combination of educational resources is not possible without infringing licenses. Thus, given a course sketch and a set of licensed educational resources, how to guarantee to produce a course whose license is compliant with the licenses of the reused resources? The issue is to relax the course sketch goal to propose relevant, alternative, and license compatible educational resources to be combined in a course.

Ontology-based relaxation allows seeking alternative solutions to expand the scope of a query [2,3]. In [4], we propose a license-aware query processing strategy for distributed queries in the Web of Data. Our contribution allows us to detect and prevent license conflicts during distributed query processing. But, in the context of educational resources, several issues arise, for instance, (1) how semantically define a query from a course sketch, (2) how to define a ranking strategy of matching educational resources, and (3) how to guarantee a result set with a minimal number of pertinent educational resources.


The objective of this PhD thesis is to propose a query processing strategy to explore a knowledge graph of educational resources. In particular, the following challenges will be leveraged.

  • Defining a complex SPARQL query from a course sketch containing join, union, filter, optional operators, etc.
  • Defining a ranking strategy that, based on the enrichment of the educational resources, will provide an ordered set of relevant resources for a course sketch.
  • Defining a query relaxation strategy that guarantees a minimal number of relevant and license compatible educational resources. Ontology-based relaxation will be used to expand the scope of the query goals. 

Contributions will be validated experimentally and published on high-quality international conferences and workshops. 


[1] B. Moreau, P. Serrano-Alvarado, M. Perrin, and E. Desmontils. Modelling the Compatibility of Licenses. In the European Semantic Web Conference (ESWC), 2019.

[2] G. Fokou, S. Jean, A. Hadjali, and M. Baron. RDF Query Relaxation Strategies Based on Failure Causes. In the European semantic web conference (ESWC), 2016.

[3] H. Huang, C. Liu, and X. Zhou. Approximating query answering on RDF databases. World Wide Web, 15(1), 2012.

[4] B. Moreau and P. Serrano-Alvarado. Ensuring License Compliance in Linked Data with Query Relaxation. In Transactions on Large-Scale Data- and Knowledge-Centered System, volume 12920. Sept. 2021.

Copyright : LS2N 2017 - Mentions Légales -