Abstract:

This article introduces JudoMaths, a competency-based learning support system for the teaching of mathematics in cycle 3 of the French education system. Using semantic and artificial intelligence technologies, JudoMaths aims to personalize the learner's educational journey and assist the teacher in various recurring tasks. The system is based on a semantic reasoner, ontologies, and knowledge bases developed specifically to guide the learner in the acquisition of mathematical skills, in accordance with the framework established by Eduscol for cycle 3. This initiative not only responds to a need expressed by teachers but also seeks to improve student success and monitoring while lightening the teacher's burden for certain automatable tasks. The semantic aspect facilitates a formal description of mathematical skills and improves the interpretation and interoperability of the knowledge used by the system, despite certain identified ethical, technological, and behavioral limits.

The development of JudoMaths is part of a desire to intelligently integrate AI and semantic technologies into education, responding to the regression observed in the level of mathematics in France and to discussions with several primary school teachers. The system offers personalized support to students by adapting the exercises and concepts worked on according to their level and progress, thus improving engagement and information retention, while providing teachers with data visualization and automation tools. tasks that are not very educationally rewarding.

The JudoMaths architecture is based on ontologies and knowledge bases using the OWL language for a formalized and structured representation of mathematical knowledge, facilitating logical reasoning and adaptation of the system to the needs of the learner. The system, developed with Apache Jena, offers a user interface divided into two parts: one for the teacher, allowing the creation and modification of student profiles and the addition of educational content, and one for the student, offering monitoring personalized development and the possibility of engaging in adapted educational activities.

The benefits of JudoMaths include a reduction in time-consuming tasks for the teacher, better monitoring of student progress, and a skills approach that links studied concepts and practical application, while promoting the development of transversal skills. Nevertheless, the system must be used as a complement to traditional courses, requiring careful adaptation and integration into educational units. Ethical questions also emerge regarding the management of sensitive data, essential to the relevant functioning of the system but requiring particular attention.

In conclusion, JudoMaths represents a advancement in the application of AI and semantic technologies to education, offering a promising tool for personalizing learning and supporting the teacher, while emphasizing the importance of maintaining human interaction in education.

Keywords:

Personalized learning, Artificial Intelligence, Semantic Technologies, Educational support.