Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

On conceptualisation and an overview of learning path recommender systems in e-learning

A. Fuster-López, J. M. Cruz, P. Guerrero-García, E. M. T. Hendrix, A. Košir, I. Nowak, L. Oneto, S. Sirmakessis, M. F. Pacheco, F. P. Fernandes, A. I. Pereira

TL;DR

The paper addresses personalization of learning paths in e-learning by proposing a cohesive framework and a set of five recommender approaches within the iMath project. It develops and compares methods ranging from a top-layer concept-map/graph-walk system to bottom-layer data-driven approaches (collaborative filtering, clustering, supervised learning, and reinforcement learning) and discusses their integration. Key contributions include a two-layer architecture that couples conceptual planning with granular, data-driven recommendations, and preliminary experiments illustrating the potential synergy between layers. The work has practical impact for scalable, personalized e-learning in mathematics platforms like MathE, motivating future systematic experiments and performance-driven evaluations.

Abstract

The use of e-learning systems has a long tradition, where students can study online helped by a system. In this context, the use of recommender systems is relatively new. In our research project, we investigated various ways to create a recommender system. They all aim at facilitating the learning and understanding of a student. We present a common concept of the learning path and its learning indicators and embed 5 different recommenders in this context.

On conceptualisation and an overview of learning path recommender systems in e-learning

TL;DR

The paper addresses personalization of learning paths in e-learning by proposing a cohesive framework and a set of five recommender approaches within the iMath project. It develops and compares methods ranging from a top-layer concept-map/graph-walk system to bottom-layer data-driven approaches (collaborative filtering, clustering, supervised learning, and reinforcement learning) and discusses their integration. Key contributions include a two-layer architecture that couples conceptual planning with granular, data-driven recommendations, and preliminary experiments illustrating the potential synergy between layers. The work has practical impact for scalable, personalized e-learning in mathematics platforms like MathE, motivating future systematic experiments and performance-driven evaluations.

Abstract

The use of e-learning systems has a long tradition, where students can study online helped by a system. In this context, the use of recommender systems is relatively new. In our research project, we investigated various ways to create a recommender system. They all aim at facilitating the learning and understanding of a student. We present a common concept of the learning path and its learning indicators and embed 5 different recommenders in this context.
Paper Structure (29 sections, 3 figures, 2 tables)

This paper contains 29 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Problem formulation
  3. Conceptual problem formulation
  4. Scope of the tool
  5. Explanatory example
  6. Related work
  7. Developed methods
  8. Concept map and graph walk-based method
  9. Basic approach
  10. Input and output
  11. Advantages and drawbacks
  12. Collaborative filtering recommender
  13. Basic Approach
  14. Input and output
  15. Advantages and drawbacks
  16. ...and 14 more sections

Figures (3)

  • Figure 1: System overview
  • Figure 2: An example of the concept map. Arcs are directed and weighted. The size of the node is proportional to the number of recommended items this concept involves.
  • Figure 3: System interoperability