Open Access

Downloads

Download data is not yet available.

Abstract

Introduction: Major selection is always a matter of concern for students who have just graduated from high school and parents who have children to go to universities. Currently, there are many students who selected the wrong major, leading to unexpected learning results and wasting time and money. In fact, many students do not know which majors they are suitable for. The paper proposes a model of decision-making support in choosing majors for students immediately after graduating from high school using deep learning.


Methods: The model applied the XGBoost algorithm to build a decision tree for classification, mining educational data from which the student's ability and learning propensity are predicted and the appropriate majors are suggested.


Results: The data used for the system are collected from 1709 students’ results at the high school, the survey results on personal interests and personality, the teacher's comments and the results on major selection after graduation. From these data, the authors have built a model to advise students choosing the right major to continue their higher education.


Conclusion: The model is evaluated and verified through actual experiments with a high accuracy of 86% and proves the contribution of deep learning models to education.



Author's Affiliation
Article Details

Issue: Vol 26 No 2 (2023)
Page No.: 2828-2837
Published: Jul 31, 2023
Section: ENGINEERING AND TECHNOLOGY
DOI: https://doi.org/10.32508/stdj.v26i2.4087

 Copyright Info

Creative Commons License

Copyright: The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License CC-BY 4.0., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

 How to Cite
Huynh, K. T., & Ly, N. (2023). A deep learning model of major consulting support. VNUHCM Journal of Science and Technology Development, 26(2), 2828-2837. https://doi.org/https://doi.org/10.32508/stdj.v26i2.4087

 Cited by



Article level Metrics by Paperbuzz/Impactstory
Article level Metrics by Altmetrics

 Article Statistics
HTML = 331 times
PDF   = 85 times
Total   = 85 times