Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal KOMPUTIKA - Jurnal Sistem Komputer
Prasetyo Nugroho3, Eddy
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

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

Found 1 Documents
Search

 1 
Parallel Computing pada Clustering K-Means untuk Analisis Keketatan Program Studi SNBT 2023 Firdaus, Alif Faturahman; Fahriza Fitriani, Azzahra; Prasetyo Nugroho3, Eddy
Komputika : Jurnal Sistem Komputer Vol. 14 No. 1 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i1.14983

Abstract

This study aims to analyze the competitiveness of study programs in the 2023 SNBT dataset using the Knowledge Discovery in Databases (KDD) method and the K-Means Clustering algorithm. The competitiveness of study programs is measured by the ratio between the number of applicants and available slots, reflecting the level of competition and popularity of the programs. Two main issues are addressed: the urgency of data-driven decision-making for formulating effective student admission policies and the lengthy execution time on large datasets such as the 2023 SNBT data, which includes thousands of study programs with complex variables. The number of clusters was determined using the elbow method, dividing the data into three categories: low, medium, and high. Clustering evaluation was conducted using the silhouette score metric, revealing that Cluster 0 (low) demonstrated the best quality with the highest silhouette score. To accelerate the analysis process, parallel computing was implemented using the joblib, scikit learn and multiprocessing library, significantly reducing execution time compared to conventional methods. With an average silhouette score of 0.684816, the results indicate good clustering quality. These findings provide valuable insights for universities in understanding the competitiveness patterns of study programs and support the development of more effective and efficient data-driven student admission policies.
Co-Authors Fahriza Fitriani, Azzahra Firdaus, Alif Faturahman