Garuda - Garba Rujukan Digital

YOLOv11-Based Automated PPE Detection System for Workplace Safety Monitoring in Electric Power Distribution Operations Ordrick, Jevon; Wibowo, Galih Hendra; Fahmi, Arif; Kurniawan, Indra; Haq, Endi Sailul
Journal of Information System and Informatics Vol 7 No 4 (2025): December
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v7i4.1379

Manual monitoring of Personal Protective Equipment (PPE) compliance in electric power distribution is prone to human error, limited supervision, and geographically dispersed work sites. This study proposes an automated PPE detection system using the YOLOv11 deep learning model to enhance safety monitoring at PT PLN (Persero) UP3 Banyuwangi. A dataset of 589 images containing 1,425 labeled PPE instances across seven categories was used to train the YOLOv11s model. The system was deployed via a web-based application with adjustable detection thresholds and validated through interviews with three OHS supervisors. It achieved 94.0% precision, 90.1% recall, and 92.8% mAP@50, with perfect detection for persons and near-perfect results for full-body harnesses. The application processed images in 2–3 seconds on standard CPU hardware, supporting automated documentation for compliance reporting. This is the first known YOLOv11-based PPE detection system tailored to electric power distribution settings. While results are promising, limitations include a small validation set and lower accuracy in detecting safety boots. Future work should explore real-time video analysis, system integration, and long-term studies on safety compliance improvements.

Utilization of the AgriTrack Information System to Strengthen Smart Farming Practices in Small-Scale Hydroponic Enterprises Sholeha, Eka Wahyu; Supriyanto, Arif; Utomo, Hendrik Setyo; Firmansyah, Eka Ridhoni August; Aisyah, Aisyah; Hidayat, Ardhi; Mardhiyatirrahmah, Liny
Journal of Information System and Informatics Vol 7 No 4 (2025): December
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v7i4.1385

The implementation of smart farming in small-scale hydroponic enterprises is often constrained by high automation costs and technological complexity. This study examines the utilization of the AgriTrack information system as a practical approach to strengthening smart farming practices through structured digital data management. AgriTrack was utilized in a small-scale hydroponic farm using a Software Development Life Cycle (SDLC) Waterfall approach, encompassing system configuration, operational deployment, and evaluation through functional testing and user acceptance testing. The system applies a cycle-based relational data model to manage cultivation records from sowing to harvesting and integrates automated scheduling with Telegram Bot notifications. Testing results indicate a 100% success rate across core operational functions, while user evaluation shows that routine cultivation data recording time was reduced from several minutes to under one minute per entry. Notification delivery was consistently observed within approximately one minute after scheduled triggers, supporting timely operational decisions. These findings demonstrate that AgriTrack effectively strengthens smart farming practices in MSME-scale hydroponic enterprises by improving efficiency and accountability, while providing a scalable foundation for gradual adoption of advanced technologies such as IoT and data analytics.

Comparative Analysis of Random Forest, Logistic Regression and SVM for Stunting Prediction Using Anthropometric Data Widyawati, Shalsa Bela Dwi; Purwadi, Purwadi; Yunita, Ika Romadoni
Journal of Information System and Informatics Vol 7 No 4 (2025): December
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v7i4.1387

Stunting remains a critical nutritional issue in Indonesia, significantly impacting the physical and cognitive development of children under five. Prompt and accurate detection of nutritional status is essential for early intervention. This study aims to predict toddlers' nutritional health using the Random Forest algorithm, based on age and height data. From an initial dataset of 120,998 anthropometric records, preprocessing steps—such as duplicate removal and nutritional status recategorization—resulted in a final dataset of 39,425 entries. The research methodology includes data collection, preprocessing, exploratory analysis, model training, handling class imbalance, and performance evaluation using accuracy, precision, recall, and F1-score. The study also compares the Random Forest model with Logistic Regression and Support Vector Machine (SVM). Results show that Random Forest outperforms the other models, achieving perfect classification metrics: Accuracy (1.00), Recall (1.00), F1-Score (1.00), and Cross-validation Accuracy (99.74%). These outcomes highlight Random Forest's robustness in classifying under-five nutrition data, making it an effective tool for rapid and reliable stunting risk detection. This research supports efforts to reduce Indonesia's stunting rate to below 20% by 2024, contributing to national health improvement strategies through technology-driven early diagnosis.

Sentiment Analysis of User Reviews for AI Applications: Evaluating SVM, Logistic Regression, and Random Forest Rivana Dwi Cahyani; Putri Taqwa Prasetyaningrum
Journal of Information System and Informatics Vol 8 No 1 (2026): February
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i1.1366

The rapid growth of AI applications such as CICI, GROK, and Gemini has resulted in a large volume of user reviews on platforms like the Google Play Store, making sentiment analysis a critical tool for understanding user perceptions. This study compares the performance of three machine learning models: Random Forest, Support Vector Machine (SVM), and Logistic Regression in classifying sentiments in 3,500 Indonesian-language reviews. A hybrid feature extraction approach, combining sentiment lexicons with TF-IDF, was applied to improve sentiment classification accuracy. The models were evaluated based on accuracy, precision, recall, and F1-score. Results indicated that all models achieved an accuracy greater than 96%, with Random Forest providing the most consistent and accurate results, achieving an overall accuracy of 99.62%. While SVM excelled in classifying positive and negative sentiments, it faced challenges with neutral reviews due to the ambiguity and overlap in sentiment expression. Logistic Regression also showed strong performance, especially on structured reviews. The findings suggest that Random Forest is the most robust and reliable model for sentiment analysis, particularly in handling diverse AI application reviews. These results offer practical insights for developers seeking to improve application performance by leveraging sentiment analysis on user feedback.

Implementation of the Collaborative Filtering Method for a Clothing Sales Recommendation System in Fashion Store Ifrah Ayyuna; Triase Triase
Journal of Information System and Informatics Vol 8 No 1 (2026): February
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i1.1368

The rapid growth of e-commerce has made personalized product recommendations a crucial aspect of enhancing customer satisfaction and boosting sales. However, many small-to-medium-sized retail businesses, like Adiva Fashion Store, still rely on manual product selection through customer searches or seller recommendations, which often leads to challenges in meeting customer preferences. This study presents a case study of Adiva Fashion Store, where the Collaborative Filtering method was implemented to develop a personalized clothing product recommendation system. The item-based Collaborative Filtering approach was employed to calculate the similarity between products based on customer ratings and transaction history. These similarity values were then used to predict customer preferences for products that had not yet been purchased. The system was developed using the Waterfall methodology, which involved needs analysis, system design, implementation, testing, and maintenance. The results show that the recommendation system significantly improved the relevance of product suggestions, helping customers make better purchasing decisions and increasing sales effectiveness. This case study illustrates how data-driven recommendation systems can be effectively integrated into small-to-medium-sized retail environments, providing valuable insights for other businesses aiming to adopt similar strategies.

Comparative Performance Analysis of Random Forest and Logistic Regression for Sentiment Classification of the Makan Bergizi Gratis Program on Platform X Slamet Endro Prianto; Berlilana Berlilana; Rujianto Eko Saputro
Journal of Information System and Informatics Vol 8 No 1 (2026): February
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i1.1371

The rapid growth of e-commerce has made personalized product recommendations a crucial aspect of enhancing customer satisfaction and boosting sales. However, many small-to-medium-sized retail businesses, like Adiva Fashion Store, still rely on manual product selection through customer searches or seller recommendations, which often leads to challenges in meeting customer preferences. This study presents a case study of Adiva Fashion Store, where the Collaborative Filtering method was implemented to develop a personalized clothing product recommendation system. The item-based Collaborative Filtering approach was employed to calculate the similarity between products based on customer ratings and transaction history. These similarity values were then used to predict customer preferences for products that had not yet been purchased. The system was developed using the Waterfall methodology, which involved needs analysis, system design, implementation, testing, and maintenance. The results show that the recommendation system significantly improved the relevance of product suggestions, helping customers make better purchasing decisions and increasing sales effectiveness. This case study illustrates how data-driven recommendation systems can be effectively integrated into small-to-medium-sized retail environments, providing valuable insights for other businesses aiming to adopt similar strategies.

Towards Self-Defending SDN Infrastructures: Real-Time Honeypot-Enabled Botnet Detection Using ONOS Nyamwaga M Kaare; Anael Elikana Sam
Journal of Information System and Informatics Vol 8 No 1 (2026): February
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i1.1375

Modern Software-Defined Networks (SDNs), while benefiting from centralized programmability, remain vulnerable to fast-evolving botnet attacks. This paper presents and evaluates a lightweight ONOS-based honeypot and decoy framework designed to detect and automatically block multi-vector botnet behaviors in real time. The system integrates honeypot-exposed Telnet, SMB, and DNS services with threshold-, entropy-, signature-, and correlation-based inspection within a tree topology (depth = 2, fanout = 4) consisting of five OpenFlow switches and 50 hosts. Quantitatively, the system achieved 100% detection of all signature-based attacks (55/55), 100% blocking of distributed UDP scans (50/50), and 0% false positives on benign decoy access. Median detection latency ranged between 1–3 seconds. True positives (TP), false negatives (FN), false positives (FP), and true negatives (TN) were measured using ground-truth attacker lists built into automated test scripts, yielding precision and recall of 1.00 across all malicious scenarios. This work demonstrates that combining deception with SDN-level flow automation enables effective and computationally efficient botnet defense without machine learning. A key limitation is that all evaluations were conducted exclusively in a controlled Mininet simulation, which may not fully represent real-world traffic dynamics. Future work will validate the system on physical SDN deployments and evaluate its robustness under production workloads.

Bibliometrics Analysis of Bankruptcy Prediction Trends in MSMEs: Global Insights from (2020–2025) Supriyono Supriyono; Purwanto Purwanto; Aris Sugiharto
Journal of Information System and Informatics Vol 8 No 1 (2026): February
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i1.1378

The purpose of this study is to map the development of research on bankruptcy prediction in Micro, Small, and Medium Enterprises (MSMEs) during 2020–2025 and to identify major scientific trends, influential authors, and dominant methodological approaches. Using a bibliometric method, data were collected from the Scopus database, producing 144 initial documents that were filtered into 23 final publications based on relevance and open-access availability. Performance analysis and science mapping were carried out using VOSviewer through co-authorship, co-citation, and keyword co-occurrence networks. The findings reveal four main research clusters: (1) financial-ratio-based distress models, (2) machine-learning approaches for SME risk prediction, (3) post-pandemic MSME resilience, and (4) credit scoring using non-financial indicators. Scientometrics is identified as the most influential journal, while Edward I. Altman and Alessandro Giannozzi emerge as central scholars. The United States, Italy, and the United Kingdom appear as the most collaborative and productive countries. The novelty of this research lies in its specific focus on MSME bankruptcy prediction during the post-pandemic era, the use of an open-access-filtered dataset, and the identification of emerging thematic clusters. However, this review is limited to Scopus-indexed, English-language, and open-access publications, which may exclude relevant studies from other sources.

Quantum Computing in Molecular Design and Drug Discovery: A Systematic Literature Review Charnelle Razo; Belinda Ndlovu
Journal of Information System and Informatics Vol 8 No 1 (2026): February
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i1.1380

This study examines how quantum computing (QC) is being applied to molecular design and drug discovery. This study aims to investigates how QC surpasses classical limitations, focusing on empirical performance in precision, accuracy, and optimisation tasks. Study design use PRISMA 2009 guidelines, 15 empirical studies (2020-2025) were included. Data were extracted on the drug-discovery stage, the algorithm used, evaluation metrics, benefits, and limitations. The findings show QC outperforms classical methods particularly through hybrid quantum–classical models. Thirteen studies reported superior gains, including AUC–ROC values of 0.80–0.95, +30% improvement in drug-likeness (QED), +6% increase in prediction accuracy, and up to 99% accuracy in drug–target interaction tasks. However, noisy intermediate-scale quantum (NISQ) hardware limitations and poor scalability limit real-world deployment, due to noise, and limited qubit counts. Consequently, current performance results are largely simulation-based rather than hardware-validated. In contrast to prior algorithm-centric reviews, this study provides a consolidated empirical synthesis and proposes a hybrid quantum–classical pipeline that maps high-performing algorithms across the drug discovery workflow under NISQ-era constraints. These findings inform pharmaceutical research and development by identifying realistic adoption pathways and the boundaries of current technological readiness.

A Systematic Review of Agentic AI for Threat Detection and Mitigation in 5G Networks Kudzaishe Lawal Chizengwe; Belinda Ndlovu
Journal of Information System and Informatics Vol 8 No 1 (2026): February
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i1.1382

Fifth-generation (5G) networks face escalating security challenges driven by decentralised architectures, stringent ultra-low-latency requirements, and rapidly evolving threat landscapes. Agentic Artificial Intelligence (agentic AI) autonomous systems that perceive network conditions, decide on countermeasures, and act in real time offers a promising route toward adaptive defence. This systematic review examines how agentic AI is being applied to detect and mitigate threats within 5G networks. Following PRISMA 2009 guidelines, four databases (IEEE Xplore, ACM Digital Library, SpringerLink, and ScienceDirect) were searched, yielding 22 eligible peer-reviewed studies published between 2020 and 2025, selected for explicit 5G relevance and empirical evaluation. The reviewed evidence clusters into four primary security areas: anomaly detection, DDoS mitigation, network slicing security, and intrusion detection. Across these domains, approaches based on federated learning, deep reinforcement learning, and multi-agent systems generally report stronger detection performance and/or more adaptive response behaviour than conventional, reactive baselines, while supporting privacy-preserving intelligence at the edge. However, key deployment barriers remain: 86% of studies rely on simulation-based validation, scalability beyond 100 nodes is insufficiently characterised, and reported coordination delays (120–180 ms) may conflict with 5G latency constraints in time-critical settings. To consolidate findings, this review proposes a Perception–Decision–Action–Feedback conceptual framework and highlights priorities for real-world validation and deployment-oriented evaluation.

Home Page

OAI Link

Editorial Team

Contact

Reviewer

Google Scholar

Contact Name
Usman Ependi

Contact Email
usmanependi@adsii.or.id

Phone
081271103018

Journal Mail Official
usmanependi@adsii.or.id

Editorial Address
Jl AMD, Lr. Tanjung Harapan, Taman Kavling Mandiri Sejahtera B11, Kel. Talang Jambe, Kec. Sukarami, Palembang, Provinsi Sumatera Selatan, 30151

Location
Unknown,

Unknown

INDONESIA

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Home Page

OAI Link

Editorial Team

Contact

Reviewer

Google Scholar

Contact Name Usman Ependi

Contact Email usmanependi@adsii.or.id

Phone 081271103018

Journal Mail Official usmanependi@adsii.or.id

Editorial Address Jl AMD, Lr. Tanjung Harapan, Taman Kavling Mandiri Sejahtera B11, Kel. Talang Jambe, Kec. Sukarami, Palembang, Provinsi Sumatera Selatan, 30151

Location Unknown, Unknown INDONESIA

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Contact Name
Usman Ependi

Contact Email
usmanependi@adsii.or.id

Phone
081271103018

Journal Mail Official
usmanependi@adsii.or.id

Editorial Address
Jl AMD, Lr. Tanjung Harapan, Taman Kavling Mandiri Sejahtera B11, Kel. Talang Jambe, Kec. Sukarami, Palembang, Provinsi Sumatera Selatan, 30151

Location
Unknown,

Unknown

INDONESIA