Individual Support in MINT Teaching (IGNITE)
The rapid expansion of educational programs and resources has created substantial content overlap, complicating the construction of coherent learning paths and challenging educators to address diverse student needs. This project introduces a knowledge graph–driven recommender framework that links learners’ prior knowledge to curricular components, enabling personalized course trajectories and adaptive curriculum design. By leveraging structured semantic representations, the framework reduces redundancy, enhances coherence, and fosters a more responsive educational environment.
Project Description
Significant individual differences in student knowledge and skills create challenges in meeting course requirements and achieving program-wide learning goals. Providing courses that comprehensively cover all missing knowledge is impractical due to limited teaching capacity and credit restrictions, while excessive repetition of known material reduces curriculum attractiveness. This issue is particularly relevant during the transition from diverse national and international bachelor's programs to master's studies. The project addresses this challenge through personalized teaching within modularized courses, where individual gaps are closed by tailoring compositions of module elements to each student's needs. Knowledge graphs will model both student competencies and the dependencies of module elements, enabling tools to generate optimized, individualized course paths. The project will focus on different modular courses, with experimental evaluation involving students from computer science and digital health sciences.
Project Focus
-
Personalized Learning: Generate individualized course paths based on learners’ prior knowledge and goals.
-
Adaptive Curriculum: Enable instructors to adjust content and reduce redundancy.
-
Knowledge Graphs: Use semantic representations to model data and ensure coherent learning
All At a Glance
-
Duration:
01.04.2024-31.03.2026 -
Research Area:
LLM, Digital Education -
Funding:
426.200,00€
Project Methodology
Curriculum Analysis and Database Construction:
The project collects and enriches course module metadata to link with learners’ knowledge gaps and goals. It also builds the knowledge graph database and supporting infrastructure.
Knowledge Modelling
The project builds graph models linking course modules, domains, and individual learning goals for personalized FlexModules, and provides a handbook for converting traditional courses.
Learning Personalization
The project creates personalized learning models and algorithms for FlexModule assignment, weekly path scheduling, and teacher feedback for curriculum adjustments.
Learning Assessment
The project develops processes and algorithms to identify learning gaps, align FlexModules with accreditation requirements, and create personalized assessments, integrating these into the best-practices handbook.
Project Pilots and Dissemination
The project designs blended and flipped classroom pilots, develops learning and evaluation platforms, obtains approvals, runs the pilots, collects data, and defines dissemination strategies for the results.
