Predicting Academic Performance - Toward a Model Based on Machine Learning and Learner’s Intelligences

Aug 5, 2026

Upload by Miles

Online learning generates continuous behavioral, social, emotional, and cognitive signals, making accurate academic performance prediction essential for personalization. This study proposes a competency-based, machine learning predictive model intended for integration into online learning platforms. The approach combines explicit traces (e.g., demographic information) with implicit traces capturing learner interactions. Collected learning traces are processed and analyzed to improve prediction accuracy while providing insights into skill acquisition and learners’ personal development, enabling more adaptive educational support.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

1. Introduction

2. Method

2.1. Literature review analysis

Predicting academic performance: toward a model based on machine learning and learner’s intelligences Jamal Eddine Rafiq1, Zakrani Abdelali1, Mohammed Amraouy2, Said Nouh3, Abdellah Bennane4 1Laboratory of Artificial Intelligence and Complex Systems Engineering, Hassan II University, Casablanca, Morocco 2National Institute of Posts and Telecommunications, Mohammed V University, Rabat, Morocco 3Information Technology and Modeling, Hassan II University, Casablanca, Morocco 4Inspectors Training Center for Education, Mohammed V University, Rabat, Morocco | Article history:
Received May 19, 2024 Revised Aug 20, 2024 Accepted Sep 3, 2024 | With the rapid evolution of online learning environments, the ability to predict students' academic performance has become crucial for personalizing and enhancing the educational experience. In this article, we present a predictive model based on machine learning techniques, designed to be integrated into online learning platforms using the competency-based approach. This model leverages features from four key dimensions: demographic, social, emotional, and cognitive, to accurately predict learners'academic performance. We detail the methodology for collecting and processing learning traces, distinguishing between explicit traces, such as demographic data, and implicit traces, which capture learners' interactions and behaviors during their learning process. The analysis of these data not only improves the accuracy of performance predictions but also provides valuable insights into skill acquisition and learners' personal development. The results of this study demonstrate the potential of this model to transform online education by making it more adaptive and focused on individual learners' needs.
This is an open access article under the CC BY-SA license.
| | --- | --- | | Keywords:
Competency-based learning
Digital learning Learning traces Machine learning Predicting academic Performance | | Corresponding Author: Jamal Eddine Rafiq Laboratory of Artificial Intelligence and Complex Systems Engineering, Hassan II University 150 Avenue Nile Sidi Othman, Casablanca 20670, Morocco Email: jamal.rafiq-etu@etu.univh2c.ma Article Info ABSTRACT 1. INTRODUCTION Digital learning has revolutionized education by offering new modalities, including online instruction. Despite these advances, it remains a crucial challenge to anticipate and effectively improve learners' academic performance using data generated by online learning platforms. Accurate prediction of academic performance is essential to meet individual educational needs and to adapt pedagogical strategies [1], [2] . Previous research has explored various methods to predict academic performance. Traditional approaches, such as those described in studies [3], [4], [5], primarily focus on quantitative measures like test scores. Although these methods have provided valuable insights, they often overlook the broader context of learning, including knowledge acquisition and skill development. The main contributors in this field have established fundamental techniques, but have not fully addressed the complexities of integrating multimodal data in online learning environments. Current models often struggle to provide a comprehensive prediction of academic performance by neglecting the multidimensional nature of online learning. In particular, there is a gap in the use of data from various dimensions such as demographic, social, emotional, and behavioral factors, which are crucial for a holistic understanding of learners' progress. Addressing this gap could significantly enhance the accuracy of predictions and provide more personalized support for learners, thus fostering better academic outcomes. In this study, we propose the academic performance prediction model based on competency-based learning traces (4I-CBT) . Unlike traditional methods, 4I-CBT utilizes multimodal data to provide a more precise and nuanced pre

FAQs

What is the main goal of the proposed study?

The study aims to predict learners’ academic performance in online learning settings using competency-based learning traces and machine learning.

What types of learning traces does the model use?

It leverages explicit traces such as demographic data and implicit traces that reflect learners’ interactions and behaviors during learning.

Which dimensions are used to characterize learners’ signals?

The model analyzes features across four dimensions: demographic, social, emotional, and cognitive.

Aug 5, 2026

Upload by Miles