Personalized Learning Paths: Bayesian Network-Based Pre-Assessment Grouping
Mary Gaceri Asunta, Ronald Waweru Mwangi, Michael Waema Kimwele · 2025
This experimental study aimed at improving the learning process through achieving personalized learning for efficient Intelligent Tutoring Systems (ITS) through utilizing Bayesian Networks (BN) This probabilistic approach was utilized to group learners into Beginner, Intermediate and Advanced knowledge tiers based on the learner's biographical data and results to a pre-assessment quiz carried out at the onset of the study. This was then utilized to customize instruction strategies to suit individual learner according to their proficiency group thus effectively addressing cold-start challenges that often arise in online learning environments. Bayesian Networks excel in handling uncertainty and modeling complex interactions, allowing for the accurate prediction of optimal learning paths as showcased in this study. The findings confirm that engaging learners through personalized pathways enhances their experience and validates BN as a powerful tool for adaptive learning systems, making it an effective approach to learner grouping and customized instruction.