Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Sebatik
Sarjon Defit
Magister Teknik Informatika, Universitas Putra Indonesia YPTK Padang
Computer Science & IT

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

Found 2 Documents
Search

 1 
An Analysis of Public Satisfaction with Government Services: A Multi-Method Approach Using PCA, K-Means Clustering, and Linear Regression Abuzar Gafari; Sarjon Defit; Rini Sovia
Sebatik Vol. 30 No. 1 (2026): June 2026
Publisher : STMIK Widya Cipta Dharma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46984/pcdj8z43

Abstract

Flawless performance evaluation results across all service dimensions may potentially obscure the identification of areas for improvement and diminish objectivity in decision-making. This study aims to identify the specific service attributes influencing public satisfaction and to segment respondents based on their satisfaction levels at the Office of the Ministry of Religious Affairs in Payakumbuh City. The research integrates Principal Component Analysis (PCA), K-means clustering, and linear regression. PCA was employed to reduce data dimensionality and establish principal components; K-means clustering was utilized to group respondents based on perceptual similarities regarding service quality; and linear regression was applied to identify the most significant factors influencing public satisfaction within each segment. The data were sourced from the Public Service Survey Information System (SISULAP) application of the Payakumbuh Ministry of Religious Affairs, spanning June 2024 to October 2025, with a total of 1,950 respondents. The findings reveal that service process and efficiency are the primary factors influencing all respondent segments, with the low-satisfaction segment identified as the top priority for service improvement. The regression models demonstrate robust performance across all segments. These findings provide an empirical foundation for data-driven policymaking to enhance public service quality.
Model Interpretation for Student Major Selection Using Principal Component Analysis and Random Forest Antoni Antoni; Sarjon Defit; Yuhandri Yuhandri
Sebatik Vol. 30 No. 1 (2026): June 2026
Publisher : STMIK Widya Cipta Dharma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46984/zyc1hh02

Abstract

The development of information technology has had a significant impact on the education sector by providing data-driven tools to support the process of major selection. This process often causes confusion among students due to its crucial role in determining their academic and career futures. This study aims to develop an accurate and transparent recommendation system for major selection through the integration of Principal Component Analysis (PCA), Random Forest (RF), and SHAP. The research follows a systematic framework that includes data processing and model evaluation stages. PCA is applied to reduce the dimensionality of complex student data in order to improve computational efficiency and minimize information redundancy. Furthermore, the Random Forest algorithm is employed as a classification model to predict major recommendations such as Science, Social Sciences, and Religious Studies. The SHAP method is integrated to provide both mathematical and visual interpretations of the contribution of each academic feature to the model’s prediction results. The research data are obtained from the internal records of MAN 1 Payakumbuh covering the last three academic years (2022/2023–2024/2025). The dataset consists of 571 eleventh-grade students with tenth-grade academic scores and non-academic skill variables. The implementation of this model is able to provide more objective recommendations compared to conventional subjective assessments, achieving an accuracy of 88.70%. Visualization of feature contributions using SHAP enhances transparency and facilitates stakeholders’ understanding of the basis for each model decision. This study contributes to improving the efficiency of the major selection process and supports more accurate academic decision-making for students and educators.
Co-Authors Abuzar Gafari Antoni Antoni Sovia, Rini Yuhandri Yuhandri