Open Ph.D. Position: Semantic Analysis of activities in a learning environment
Title: Semantic Analysis of activities in a learning environment: Detection of learning communities and exploitation for the recommendation of pedagogical resources, the adaptation of learning paths and the follow-up of learners
Learning ways changed and are still changing with the arrival of digital technology in all the fields of education. The Educlever Company positions itself in this dynamics by developing "smart" tools to help education stakeholders, adapting learning paths to learners and facilitate the personalization of learning paths.
The subject of this thesis is part of a research and development project of Educlever, initiated within the framework of the project "Cartographie des savoirs" (Mappping Knowledge) and aiming the development of a learning environment that (1) enables stakeholders in education to be more efficient in supporting learners and teachers, (2) provides teachers with tools for pedagogical design or redesign and for assistance in learner monitoring and (3) offers learners personalized trails built in real time according to their activities and profile. This requires detailed knowledge of the cognitive profile of each learner.
One of the results of the project "Cartographie des savoirs" has been the production of an ontological representation of a praxeological system or skills repository for the Primary level. In the framework of this thesis, we propose to go further in the integration of knowledge at play in a learning environment, by focusing on capturing a fine-grained knowledge of the profile of each learner, on modeling this knowledge based on the skill ontology and on reasoning on this ontology, which is required for the development of the targeted learning environment. Specifically, we plan to develop a system prototype to support the monitoring of learners, in the form of a set of decision support tools for teachers and a prototype engine for the dynamic construction of personalized learning paths.
The general research question addressed in this thesis is “How to take into account the cognitive and social user profile of a learner to personalize her follow-up, to recommend her pedagogical resources, to build her pedagogical path?”.
This raises two sub-questions: “How to capture and model the cognitive and social profile of a learner in a learning environment?” and “Which kinds of reasoning on the learner profile enable the personalization of the recommendation of resources and the adaptation of pedagogical path?”.
This research will be carried out in partnership between the company Educlever and the research team Wimmics, a joined team between the I3S laboratory of University Nice Sophia Antipolis and Inria Sophia Antipolis Méditerranée. The PhD student will have an office both in the premises of Educlever and those of Wimmics and will divide his time between the two environments.
[Dehors, 2007] Exploiting Semantic Web and Knowledge Management Technologies for E-learning, PhD thesis, Université Nice Sophia Antipolis, 2007
[Yessad, 2009] Construction d’un Environnement Pédagogique Adaptatif basé sur les Modèles et Techniques du Web Sémantique, PhD thesis, Université Badji Mokhtar, 2009
[Ereteo, 2011] Analyse sémantique des réseaux sociaux, PhD thesis, Telecom ParisTech, 2011
[Meng et al. 2014] Zide Meng, Fabien L. Gandon, Catherine Faron-Zucker, Ge Song: Empirical study on overlapping community detection in question and answer sites. IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2014
The student should have a master degree in computer science and a strong knowledge of web and preferably semantic web standards.
Wimmics: Catherine Faron Zucker email@example.com & Fabien Gandon firstname.lastname@example.org
Educlever: Fabrice de Comarmond email@example.com & Laurent Janin firstname.lastname@example.org