Article Per Year (5 Year)
p-Index From 2021 - 2026
4.185
P-Index
This Author published in this journals
All Journal Jurnal Teknik Informatika C.I.T. Medicom Journal of Intelligent Decision Support System (IDSS) Community Development Journal: Jurnal Pengabdian Masyarakat Jurnal Studi Islam Lintas Negara (Journal of Cross-Border Islamic Studies) International Journal of Basic and Applied Science Jurnal Mandiri IT TRIMAS: Jurnal Inovasi dan Pengabdian Kepada Masyarakat Sahabat Sosial: Jurnal Pengabdian Masyarakat West Science Business and Management International Journal of Health Sciences Jurnal Sistem Informasi Galuh Jurnal Ilmiah Research Student Jurnal Pengabdian Masyarakat Nauli Jurnal TNI Angkatan Udara Journal of Defense Technology and Engineering
Hondor Saragih
Universitas Pertahanan Republik Indonesia
Religion Aerospace Engineering Agriculture, Biological Sciences & Forestry Humanities Automotive Engineering Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Dentistry Earth & Planetary Sciences Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Environmental Science Health Professions Industrial & Manufacturing Engineering Law, Crime, Criminology & Criminal Justice Library & Information Science Mathematics Mechanical Engineering Medicine & Pharmacology Nursing Physics Public Health Social Sciences Transportation

Published : 31 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 31 Documents
Search

 1   2   3   4 
Optimizing the performance of the K-Nearest Neighbors algorithm using grid search and feature scaling to improve data classification accuracy Manurung, Jonson; Saragih, Hondor; Prabukusumo, Muhammad Azhar; Firdaus, Eryan Ahmad
Jurnal Mandiri IT Vol. 14 No. 2 (2025): Computer Science and Field
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/mandiri.v14i2.466

Abstract

The performance of distance-based classification algorithms such as K-Nearest Neighbors (KNN) is highly dependent on proper feature scaling and optimal parameter selection. Without systematic optimization, KNN may experience decreased accuracy due to feature scale disparities and suboptimal k-values. This study aims to enhance the performance of the KNN algorithm through the integration of Feature Scaling and Grid Search Cross-Validation as a parameter optimization strategy. The research employs the Breast Cancer Wisconsin Dataset, divided into 80% training and 20% testing data. Feature normalization was performed using StandardScaler, while Grid Search was applied to determine the optimal combination of parameters, including the number of neighbors (k), weighting function (weights), and distance metric (metric). The optimized KNN configuration with k = 9, weights = distance, and metric = manhattan achieved an average accuracy of 97.19%, outperforming the baseline accuracy of 93.86%. A paired t-test confirmed that the improvement was statistically significant (p < 0.05). These findings demonstrate that the synergy between feature scaling and parameter tuning can substantially improve both the accuracy and stability of KNN models. The scientific novelty of this study lies in the systematic integration of normalization and parameter optimization through Grid Search, providing an empirical framework that enhances KNN’s robustness across datasets with heterogeneous feature distributions. The proposed approach is recommended for medical data classification and can be adapted to other domains with heterogeneous numerical feature distributions.
Optimization of XGBoost hyperparameters using grid search and random search for credit card default prediction Firdaus, Eryan Ahmad; Manurung, Jonson; Saragih, Hondor; Prabukusumo, Muhammad Azhar
Jurnal Mandiri IT Vol. 14 No. 2 (2025): Computer Science and Field
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/mandiri.v14i2.468

Abstract

This study explores the optimization of the Extreme Gradient Boosting (XGBoost) algorithm for credit card default prediction through systematic hyperparameter tuning using Grid Search and Random Search methodologies. Utilizing the publicly available Default of Credit Card Clients dataset from the UCI Machine Learning Repository, the research focuses on enhancing model performance by fine-tuning critical parameters such as learning rate, maximum tree depth, number of estimators, subsample ratio, and column sampling rate. The baseline XGBoost model achieved an accuracy of 0.8118, while the tuned models using Grid Search and Random Search improved the accuracy to 0.8183 and 0.8188, respectively. Although the improvement appears modest, the optimized models exhibited enhanced balance between precision and recall, particularly in identifying defaulters within an imbalanced dataset—an essential aspect in credit risk assessment. The results demonstrate that systematic hyperparameter optimization not only improves predictive performance but also contributes to model stability and generalization. Moreover, Random Search proved to be more computationally efficient, achieving near-optimal performance with fewer evaluations than Grid Search, thereby emphasizing its practicality for large-scale financial risk modeling applications. The novelty of this study lies in the comparative evaluation of two optimization techniques within the context of financial risk prediction, providing practical insights into how efficient hyperparameter tuning can enhance the reliability and scalability of machine learning models used in real-world credit risk management systems.
Hyperparameter optimization of graph neural networks for predicting complex network dynamics using bayesian meta-learning Saragih, Hondor; Manurung, Jonson; Prabukusumo, Muhammad Azhar; Firdaus, Eryan Ahmad
Jurnal Mandiri IT Vol. 14 No. 2 (2025): Computer Science and Field
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/mandiri.v14i2.469

Abstract

The rapid growth of graph-structured data in domains such as transportation, social networks, and biological systems has increased the demand for more adaptive and efficient Graph Neural Network (GNN) architectures. However, GNN performance remains highly sensitive to hyperparameter configurations, which are often tuned through computationally expensive manual or heuristic methods. This study proposes a novel Bayesian Meta-Learning (BML)-based framework for hyperparameter optimization of GNNs aimed at improving the prediction accuracy of complex network dynamics. The framework integrates Bayesian optimization with a meta-learning prior adaptation mechanism, enabling the model to learn optimal hyperparameter distributions across multiple graph tasks. Experimental evaluations conducted on three benchmark datasets—Cora, Citeseer, and PubMed—comprising up to 20,000 nodes with diverse structural complexities, demonstrate that the proposed BML-GNN framework achieves faster convergence, lower validation loss, and higher predictive accuracy than both baseline GNN and traditional Bayesian Optimization approaches. Quantitatively, the BML-GNN model attains an R² score exceeding 0.97 with a significant reduction in RMSE, confirming its strong generalization capability. Although the method shows notable performance improvements, its computational overhead during meta-training and reliance on well-defined prior distributions represent potential limitations. Overall, the integration of Bayesian Meta-Learning provides a robust, scalable, and uncertainty-aware optimization strategy that advances the development of reliable GNN models for complex network modeling and intelligent system design.
Swarm driven automatic feature selection and classification framework for parkinson voice data Prabukusumo, Muhammad Azhar; Saragih, Hondor; Manurung, Jonson
Jurnal Mandiri IT Vol. 14 No. 2 (2025): Computer Science and Field
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/mandiri.v14i2.470

Abstract

Parkinson’s disease (PD) severely impairs motor and vocal functions, and early detection is crucial for effective intervention. Conventional diagnostic procedures remain subjective and time-consuming, highlighting the need for automated, data-driven approaches. This study aims to develop an intelligent and fully automated framework integrating Particle Swarm Optimization (PSO)–based feature selection with ensemble machine learning classifiers for PD detection using voice data. The proposed Swarm-Driven Automatic Feature Selection and Classification Framework (SAFSCF) automates data preprocessing, adaptive feature optimization, and classification within a unified pipeline. The framework was evaluated on the Parkinson’s Speech Dataset comprising 743 numerical features. Baseline models achieved accuracies of 0.7738 (Logistic Regression), 0.8651 (Random Forest), and 0.8690 (Gradient Boosting). After PSO optimization, the feature set was reduced by nearly 50% to 382 attributes, achieving a test accuracy of 0.8421 slightly higher than the full-feature model (0.8355). Convergence plots confirmed that PSO effectively minimized the fitness function while maintaining high classification stability. Feature importance analysis revealed that the most discriminative attributes were derived from log energy, Teager Kaiser energy operators (TKEO), MFCCs, Shimmer, and entropy-based features biomarkers known to reflect Parkinsonian speech degradation. These findings demonstrate that the proposed framework enhances computational efficiency and interpretability, offering a reproducible and scalable solution for non-invasive, voice-based PD diagnosis.
Web-based development of room management information system at Universitas Pertahanan using Rapid Application Development Anjani, Prasashti Alya; Saragih, Hondor; Hidayati, Ajeng; Anindito
Journal of Intelligent Decision Support System (IDSS) Vol 7 No 3 (2024): Intelligent Decision Support System (IDSS)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/idss.v7i3.254

Abstract

Unhan RI is an educational institution responsible for facilitating the continuity of student’s academic activites, including the scheduling process that managed by the department’s staff. The scheduling process requires components such as courses, lectures, time slots, and the classrooms. The number of available classrooms at Unhan RI is less than it need. Therefore, a proper scheduling system is necessary to manage scheduling and avoid conflicts between schedule. The development of information management system for administration’s process that are still done manually are needed in this digital era. Because the large and continuously growing amount of data is difficult to process manually. The development is using Rapid Application Development method. This method is chosen because of the requirement time for the developing is short.  By using the room management information system, the process of scheduling courses and managing rooms can be done easily. This system provides information of room availability and ongoing activites, helping to prevent scheduling conflicts.
Comparison of Naïve Bayes Classifier and Support Vector Machine for sentiment analysis on civil military relations conflict among Rohingya refugees as recommendation for defense policy making Putri, Nanda Selviana; Saragih, Hondor; Heikhmakhtiar, Aulia Khamas
Journal of Intelligent Decision Support System (IDSS) Vol 7 No 3 (2024): Intelligent Decision Support System (IDSS)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/idss.v7i3.255

Abstract

This research focuses on the evaluating the performance of various sentiment analysis techniques using the Naive Bayes Classifier and Support Vector Machine in identifying civil-military conflicts among Rohingya refugees. The goal is to assist leaders in formulating defense policies. This research uses text data from news sources on Twitter, with a total of 5018 data that have been processed to become clean data, then divided into 1004 test data and 4018 training data to be classified using the Support Vector Machine and Naive Bayes methods. This research analyzes the sentiment and polarity of public opinion related to the issues that occur in this situation. The results of the sentiment analysis from the two methods are then classified using the Support Vector Machine and Naive Bayes methods, and then compared to determine which method is more effective in capturing the complex dynamics of sentiment. The findings of this research indicate that the Support Vector Machine method has a higher accuracy in identifying sentiments related to the civil-military conflict among Rohingya refugees, with an accuracy of 87.95%, compared to the Naive Bayes Classifier with an accuracy of 85.16%. The analysis results in the form of frequently occurring words in the true positive word cloud, namely apology, human, angry, and solidarity, are handed over to experts to be formulated into recommendation sentences and can be used to assist in the formulation of policies for defense decision-makers in more effectively addressing the Rohingya refugee issue.
PENCEGAHAN DAN PENANGANAN STUNTING DI KAMPUNG KWUHKENDAK DISTRIK FAKFAK BARAT WILAYAH KERJA PUSKESMAS KWUHKENDAK Tawil, Muh. Risal; Erika, Erika; Parwati, Dewi; Putri, Noviyanti Rahardjo; P, Nur Triningtias; Mainassy, Meillisa Carlen; Saragih, Hondor; Pannyiwi, Rahmat
Community Development Journal : Jurnal Pengabdian Masyarakat Vol. 5 No. 3 (2024): Volume 5 No. 3 Tahun 2024
Publisher : Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/cdj.v5i3.29224

Abstract

Stunting adalah masalah gizi utama yang akan berdampak pada kehidupan sosial dan ekonomi dalam masyarakat. Ada bukti jelas bahwa individu yang stunting memiliki tingkat kematian lebih tinggi dari berbagai penyebab dan terjadinya peningkatan penyakit. Stunting akan mempengaruhi kinerja pekerjaan fisik, fungsi mental, dan intelektual akan terganggu serta berhubungan dengan gangguan fungsi kekebalan dan meningkatkan risiko kematian. Stunting merupakan permasalahan gizi yang timbul akibat kurangnya asupan nutrisi sehingga mengakibatkan gangguan pertumbuhan pada anak. Kegiatan pengabdian kepada masyarakat ini dilakukan dengan tujuan untuk meningkatkan pengetahuan ibu mengenai stunting beserta upaya pencegahannya. Metode edukasi yang digunakan yaitu melalui penyuluhan. Kegiatan dilakukan dengan tahapan pre- test, pemberian edukasi melalui penyuluhan tentang stunting beserta pecegahannya, dan terakhir adalah post-test. Analisis data dilakukan menggunakan uji Paired Samples T-Test dengan subjek yaitu ibu-ibu yang memiliki balita usia 6-24 bulan. Hasil menunjukkan bahwa terdapat peningkatan nilai rata-rata pengetahuan para ibu setelah mendapatkan edukasi yaitu sebesar 11,19 poin, nilai rata-rata pre-test = 70,27 dan nilai rata-rata post-tes=81,46. Hasil analisis statistik menunjukan nilai signifikansi sebesar 0,000 (< 0,05), sehingga dapat disimpulkan bahwa pemberian edukasi melalui penyuluhan berpengaruh terhadap peningkatan pengetahuan ibu terkait stunting beserta upaya pencegahannya.
Generative AI and multi-source intelligence for automated security triage Herris, Fhatur Robby Tanzil; Saragih, Hondor; Anindito, Anindito
Journal of Intelligent Decision Support System (IDSS) Vol 8 No 4 (2025): December: Intelligent Decision Support System (IDSS)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/idss.v8i4.326

Abstract

Security Operation Center (SOC) analysts encounter significant delays due to "Swivel Chair Analysis," a manual and fragmented process for triaging Indicators of Compromise (IoC). This study addresses this inefficiency by developing "CyberGuardianBot," an automated ChatOps assistant built using the Rapid Application Development (RAD) methodology and the Telegram Bot API. Applying Security Orchestration, Automation, and Response (SOAR) principles, the system asynchronously orchestrates multi-source intelligence from VirusTotal, AbuseIPDB, URLScan.io, AlienVault OTX, and MobSF. A key novelty is the integration of Google Gemini to perform cognitive synthesis, translating raw API data into actionable insights. Blackbox testing validated the system across 15 test cases, confirming the successful automation of URL, IP, and file triage. The bot generates natural language executive summaries and structured reports (.txt and .pdf), significantly enhancing the speed and accuracy of the triage process while reducing the cognitive load on analysts.
Blockchain-enhanced security framework for defense supply chain management: an AI-driven smart contract approach with distributed ledger technology Hondor Saragih; Jonson Manurung; Hengki Tamando Sihotang; I Made Aditya Pradhana Putra
Journal of Defense Technology and Engineering Vol. 1 No. 2 (2026): January, Journal of Defense Technology and Engineering
Publisher : Fakultas Teknik dan Teknologi Pertahanan, Universitas Pertahanan Republik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Defense supply chains face critical security challenges including counterfeit components, unauthorized access, data tampering, and supply chain attacks that compromise operational integrity and national security. Existing blockchain implementations suffer from limited scalability, inadequate threat detection mechanisms, and insufficient integration with modern AI technologies for real-time security monitoring. This research develops an AI-Enhanced Blockchain Security Framework combining smart contracts with distributed ledger technology specifically designed for defense supply chain management. The framework employs multi-signature authentication, cryptographic verification, and machine learning-based anomaly detection across a three-layer architecture (blockchain layer, security layer, analytics layer). Validation using the DataCo supply chain dataset (180K operations) and Backstabber's knife collection attack patterns (174 documented attacks) demonstrates 94.7% attack detection accuracy, 87.3% reduction in unauthorized access attempts, and 99.2% data integrity verification rate. The system achieved 850 transactions per second (TPS) throughput with 1.8-second average latency and 40% cost reduction compared to traditional centralized systems. Smart contract execution showed 99.96% reliability across 10,000 test scenarios with automated enforcement of security policies. Statistical validation confirmed significant superiority over conventional approaches (p<0.001). Future work includes quantum-resistant cryptography, federated learning for privacy-preserving analytics, cross-chain interoperability, and integration with IoT sensors for real-time supply chain monitoring.
Pengembangan Potensi Masyarakat dalam Bijak Memanfaatkan Kecerdasan Buatan (AI) untuk Masa Depan di SMAN 1 Plered Kabupaten Purwakarta Firdaus, Eryan Ahmad; Manurung, Jonson; Saragih, Hondor; Prabukusumo, Muhammad Azhar; Hidayati, Ajeng; Kartika, Amelia
Jurnal Pengabdian Masyarakat Nauli Vol. 5 No. 1 (2026): Agustus, Jurnal Pengabdian Masyarakat Nauli
Publisher : Marcha Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/nauli.v5i1.379

Abstract

Perkembangan Kecerdasan Buatan (Artificial Intelligence/AI) telah membawa perubahan signifikan dalam berbagai bidang, termasuk pendidikan. Di satu sisi, AI memberikan berbagai kemudahan dalam proses pembelajaran, peningkatan kreativitas, serta efisiensi dalam memperoleh dan mengolah informasi. Namun, pemanfaatan AI yang tidak bijak dapat menimbulkan berbagai dampak negatif, seperti ketergantungan teknologi, plagiarisme, serta ancaman terhadap privasi dan keamanan data. Oleh karena itu, diperlukan upaya peningkatan literasi digital dan pemahaman etika dalam penggunaan AI bagi generasi muda. Kegiatan Pengabdian kepada Masyarakat (PkM) ini dilaksanakan di SMAN 1 Plered Kabupaten Purwakarta dengan tema “Pengembangan Potensi Masyarakat dalam Bijak Memanfaatkan Kecerdasan Buatan (AI) untuk Masa Depan”. Kegiatan ini bertujuan untuk meningkatkan pemahaman siswa mengenai konsep dasar AI, manfaat dan risikonya, serta mendorong pemanfaatan AI secara etis dan bertanggung jawab. Metode pelaksanaan dilakukan melalui penyampaian materi, diskusi, dan sesi tanya jawab secara interaktif. Hasil kegiatan menunjukkan bahwa peserta memperoleh pemahaman yang lebih baik mengenai penggunaan AI sebagai alat pendukung pembelajaran dan pengembangan diri. Kegiatan ini diharapkan dapat membentuk generasi muda yang adaptif, kreatif, dan siap menghadapi transformasi digital di masa depan.
Co-Authors Adha, Rochedi Idul Ajeng Hidayati Anindito Anindito, Anindito Anisa Purnamasari Anjani, Prasashti Alya Arif Rachman Asmi, Syamsinar Aulia Khamas Heikmakhtiar Bartolomeu dos Reis Chan, Ricky Cut Maisyarah Karyati Dewi Parwati Dhaifullah, Rendi Hanif Erika Erika Eryan Ahmad Firdaus Febrian Wahyu Christanto Fitriana Fitriana Gandhara, Seto Harfika, Meiana Heikhmakhtiar, Aulia Khamas Hendra Manurung, Hendra Hengki Tamando Sihotang Herris, Fhatur Robby Tanzil Hidayati , Ajeng Hidayati, Ajeng Hutabarat, Erny I Made Aditya Pradhana Putra Iman Saufik Suasana Kartika, Amelia Lediwara, Nadiza Mainassy, Meilisa Carlen Makualaina, Fenska Narly Manurung, Jonson Mardamsyah, Adam Marunung, Hendra Meillisa C. Mainassy Muhadi Sugiono Muhamad Risal Tawil Muhammad Naufal Nafian Abror Mulat, Trimaya Cahya Naufal Rizki, Muhammad Nurul Aisyiyah Puspitarini P, Nur Triningtias Pannyiwi, Rahmat Prabukusumo, Muhammad Azhar Putri, Nanda Selviana Putri, Noviyanti Rahardjo Rahmat, Rezqiqah Aulia Rezasyah, Teuku Ricky Chan Sadewa, Rio Bintang Saputra, M. Khalid Fredy Satria, Tegar Pandu Sembada Denrineksa Bimorogo Shanti Maulani Simarmata, Dicky Daniel Sipahutar, Pongki Sofiana Sofiana, Sofiana Tegar Pandu Satria Treasa, Adisty Dwi Vernando, Deden Vicente Soares Zefanya Seto Gandhara