Article Per Year (5 Year)
p-Index From 2020 - 2025
0.408
P-Index
This Author published in this journals
All Journal Educenter: Jurnal Ilmiah Pendidikan EDUCENTER JURNAL PENDIDIKAN
Zaidan, Alfa Hikma
Unknown Affiliation
Economics, Econometrics & Finance Education Health Professions

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Implementation of deep learning in the recommendation system for high school admissions in Semarang City Zaidan, Alfa Hikma; Ningrum, Novita Kurnia
Educenter : Jurnal Ilmiah Pendidikan Vol. 4 No. 2 (2025): Educenter: Jurnal Ilmiah Pendidikan
Publisher : ARKA INSTITUTE

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55904/educenter.v4i2.1766

Abstract

The New Student Admission System for senior high schools in Semarang still faces challenges in achieving fair and proportional selection. The dominant zoning policy often ignores students' academic potential; therefore, a more comprehensive recommendation system is needed. This study proposes the development of a deep learning-based school recommendation system using a Multi-Layer Perceptron (MLP) architecture with a backpropagation algorithm. The dataset consists of 16 public senior high schools in Semarang, with the main variables including exam scores, age, school capacity, and distance from student residence calculated using the Euclidean distance method. The data is divided into a training set and a test set, with normalization applied to all numeric features. The training results show high accuracy. The system is able to generate school recommendation rankings that are visualized in tabular formats and interactive maps. Experimental results indicate that distance and school capacity contribute significantly to determining preference scores. Therefore, this study confirms that the deep learning approach is more adaptive than the rule-based linear method and can be an alternative solution to support a fairer and more transparent Student Admissions policy. For further research, it is recommended to develop the system by adding more diverse variables, real-time data integration, and implementing a more complex deep learning architecture to optimize the quality of recommendations.
Implementation of deep learning in the recommendation system for high school admissions in Semarang City Zaidan, Alfa Hikma; Ningrum, Novita Kurnia
Educenter : Jurnal Ilmiah Pendidikan Vol. 4 No. 2 (2025): Educenter: Jurnal Ilmiah Pendidikan
Publisher : ARKA INSTITUTE

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55904/educenter.v4i2.1766

Abstract

The New Student Admission System for senior high schools in Semarang still faces challenges in achieving fair and proportional selection. The dominant zoning policy often ignores students' academic potential; therefore, a more comprehensive recommendation system is needed. This study proposes the development of a deep learning-based school recommendation system using a Multi-Layer Perceptron (MLP) architecture with a backpropagation algorithm. The dataset consists of 16 public senior high schools in Semarang, with the main variables including exam scores, age, school capacity, and distance from student residence calculated using the Euclidean distance method. The data is divided into a training set and a test set, with normalization applied to all numeric features. The training results show high accuracy. The system is able to generate school recommendation rankings that are visualized in tabular formats and interactive maps. Experimental results indicate that distance and school capacity contribute significantly to determining preference scores. Therefore, this study confirms that the deep learning approach is more adaptive than the rule-based linear method and can be an alternative solution to support a fairer and more transparent Student Admissions policy. For further research, it is recommended to develop the system by adding more diverse variables, real-time data integration, and implementing a more complex deep learning architecture to optimize the quality of recommendations.
Co-Authors Ningrum, Novita Kurnia