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

2024-06-07

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.

Information Retrieval,Artificial Intelligence,Machine Learning

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to effectively recommend personalized learning paths in e - learning environments. Specifically, researchers are concerned with how to personalize students' online learning paths through recommendation systems to improve students' learning effectiveness and satisfaction. This involves several key aspects:

1. **Personalized Learning Paths**: Traditional learning methods usually rely on fixed learning paths set by experts, and this method may overlook the learning styles, preferences, and knowledge levels of different students. Therefore, researchers hope to develop a recommendation system that can go beyond static structures and dynamically adjust learning paths according to each student's individual situation.

2. **Evaluation and Feedback**: Compared with traditional learning paths, this study emphasizes the importance of evaluation in the recommendation system. The recommendation system not only needs to provide personalized suggestions based on students' current knowledge levels and learning progress, but also needs to continuously optimize the recommended content through continuous evaluation and feedback mechanisms.

3. **Algorithms and Technologies**: In order to achieve the above - mentioned goals, researchers have explored a variety of algorithms and technologies, including methods based on concept maps and graph traversal, collaborative filtering, cluster analysis, supervised learning, and reinforcement learning, etc. These methods aim to solve the problem of personalized learning path recommendation from different perspectives, for example, by analyzing students' behavior data to predict their learning needs, or using machine - learning algorithms to automatically generate and optimize learning paths.

4. **Integration of Educational Tools**: The study also explores how to integrate these recommendation systems into existing educational platforms, such as the MathE platform, to enhance their educational functions. In this way, not only can students' self - learning abilities be improved, but also valuable feedback can be provided for teachers to help them better understand students' learning states and needs.

In summary, the core problem of this paper is to realize the recommendation of personalized learning paths in e - learning environments by developing and applying advanced recommendation system technologies, thereby improving learning efficiency and student satisfaction.