Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal International Transactions on Artificial Intelligence (ITALIC)
Shilpa Mehta
Auckland University of Technology
Computer Science & IT Control & Systems Engineering Library & Information Science

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Playing Smart with Numbers: Predicting Student Graduation Using the Magic of Naive Bayes Shilpa Mehta
International Transactions on Artificial Intelligence Vol. 2 No. 1 (2023): International Transactions on Artificial Intelligence
Publisher : Pandawan Sejahtera Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33050/italic.v2i1.405

Abstract

The quality of a higher education institution is often measured by the accreditation granted by the National Accreditation Agency for Higher Education (BAN-PT). In this context, one of the primary assessment criteria is the graduation rate of students. An intriguing study employs the Naive Bayes algorithm to forecast whether students will graduate on time or face delays. The resulting predictive outcomes offer valuable insights and input to universities for enhancing their educational standards. The Naive Bayes method brings its unique advantages, particularly in predicting graduation rates based on real-world data. This ensures that the generated predictions can be relied upon and utilized as guidelines for future projections. This predictive mechanism encompasses 14 pivotal factors. These factors include gender, student status, age, marital status, performance across semesters 1 through 8, as well as cumulative performance, culminating with the information of whether a student passed or not. Within this study, data from 302 students of the 2018 cohort were involved. Data processing was carried out using the Python programming language within the Jupyter Notebook environment. The results unveil an impressive accuracy rate, reaching 85%. In terms of precision, the prediction for delays achieved a value of 0.42, while timely graduation prediction scored 0.95. Furthermore, the accuracy in identifying delay cases reached 0.65, compared to 0.88 accuracy for timely predictions. The f1 score for delay predictions stood at 0.51, while timely graduation predictions reached 0.91. These results illustrate that this algorithmic approach is capable of providing accurate and well-balanced insights into student graduation predictions.
Co-Authors