Insights into Factors Influencing Academic Success: An Application of Classification Models in Higher Education
DOI:
https://doi.org/10.47852/bonviewAIA32021465Keywords:
machine learning, higher education, students’ performance, correlation filter, student attributes, classification algorithmsAbstract
Understanding the factors that impact students’ achievements and failures in higher education is crucial as it enables the development of targeted interventions and support mechanisms that can enhance academic performance and foster student success. Artificial intelligence (AI) can help understand the factors that influence students’ academic achievements and failures in higher education by analyzing large volumes of data, identifying patterns and correlations, and providing valuable insights. This article presents the application of eight classification models on a “Dataset of Academic Performance Evaluation of Higher Education Students” consisting of 145 higher education students. The aim is to identify the best classification algorithm for predicting academic performance. The correlation filter (CF) was used for the discovery and selection of relevant attributes, resulting in the choice of four attributes for analysis. The best classification models were random forest, support vector machine, and decision tree, with an average accuracy of 94.37% and a CF of 0.1. These results demonstrate that the application of AI and machine learning techniques is important for decision-making in higher education, allowing for a better understanding of the factors that influence academic success or failure. The study emphasizes the importance of careful attribute selection and the use of appropriate classification algorithms to ensure accuracy and reliability of the results. Additionally, the study was replicated and evaluated with nine Brazilian higher education students, achieving an accuracy of 88.89%. These results demonstrate the consistency and relevance of the proposed attribute filtering model.
Received: 1 August 2023 | Revised: 12 October 2023 | Accepted: 20 October 2023
Conflicts of Interest
The authors declare that they have no conflicts of interest to this work.
Data Availability Statement
We are not dealing with humans in our research, as the databases used in our research are open, public, scientific database available on the UCI Machine Learning Repository website: https://archive.ics.uci.edu/dataset/856/higher+education+students+performance+evaluation. In this article, we used an open public database deposited in the repository.
Author Contribution Statement
Danielli Araújo Lima: Conceptualization, Methodology, Writing - original draft, Writing - review & editing, Visualization, Supervision, Project administration. Liliana Terrazas Balderrama: Validation, Formal analysis, Investigation, Writing - original draft.
Metrics
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.