Enhancing Personalized Learning Experiences through AI-driven Analysis of xAPI Data
Loading...
Date
2024-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The British University in Dubai (BUiD)
Abstract
In today's evolving educational arena, Adaptive learning experiences to individual needs has become a focal point. The Experience API (xAPI) provides a comprehensive mechanism to document all types of learning interactions, storing this stream of data into the Learning Record Store (LRS). This dissertation explores the fusion of Artificial Intelligence (AI) techniques with the obtained xAPI data. There is a gap of research in utilizing xAPI and AI integration in addressing learning objectives and understanding learners cognitive state and the utilization of data in actionable manner. This paper recommends a competency-aware framework for integrating xAPI and AI that predicts the pass/fail status of every student and provides personalized actionable feedback in an autonomous manner and in human-friendly language. To achieve this goal the CRISP-DM methodology was utilized. The analysis examined an eLearning lesson with 153 records and 51 participants, it concluded that blooms-level and pre-assessments are reliable predictors of student performance. The classification algorithms were able to predict the pass/fail statues with up to 93.5% accuracy. These predictions were fed to ChatGPT that provided personalized actionable feedback to students. The findings of this study can offer valuable insights for Educators, e-learning professionals, and AI researchers, showcasing the potential of AI in transforming the future of adaptive learning.
Description
Keywords
xAPI, educational data mining, AI