Proponent/Claimant
Jessie R. Paragas, Nanang Husin, Hapsari Peni Agustin Tjahyaningtijas, Lusia Rakhmawati, Rindu Puspita Wibawa, Alim Sumarno, Wahyu Pratomo Sapto Warsono, Asmunin, Hasanuddin Al-Habib
Abstract
At this moment, machine learning (ML) is used concurrently with data mining in many fields, including education. They give improved insight to identify the critical factors from the student that greatly impact student achievement and education quality. In university, research shows that early years are essential for student success in their education. Sadly, not all students can pass these critical times properly. Several indicators can indicate whether students undergo this stage well or not. Currently, the most common method to solve this problem is using an academic advisor (AA) to monitor each student. However, this method’s success rate is around 50%. To increase the success of the counseling program, especially in these early times, automated tools are built to check and predict the student’s condition. The tools will be a program with machine learning to predict the student’s graduation time and final GPA. Even though the technique used on ML is common, the implementation to provide a proper prediction for the counseling purpose gives new insight into how ML could help counseling purposes to help students succeed on their academic journey. The prediction software was designed in mind that credit load does not always correlate with student performance. It is more of a supporting variable, with the prediction weighing more on student GPA as the main variable. The prediction result will help AA to determine those who potentially could have a problem in the future so they can prepare adequately for counseling to help them make sure that they graduate with adequate GPAs and achieve better education.
Name of Research Journal
2023 International Conference on Technology, Engineering, and Computing Applications (ICTECA)
Date/Year of Publication
April 9, 2024
Citation
N. Husin et al.,
