cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
M. Miftach Fakhri
Contact Email
fakhri@unm.ac.id
Phone
+6282290603030
Journal Mail Official
wahid@unm.ac.id
Editorial Address
Program Studi Teknik Komputer, UNM Parangtambung, Daeng Tata Raya, Makassar, South Sulawesi, Indonesia
Location
Kota makassar,
Sulawesi selatan
INDONESIA
Journal of Embedded Systems, Security and Intelligent Systems
Published by Universitas Negeri Makassar
ISSN : 2745925X     EISSN : 2722273X     DOI : -
Core Subject : Science,
Computer Science & IT
The Journal of Embedded System Security and Intelligent System (JESSI), ISSN/e-ISSN 2745-925X/2722-273X covers all topics of technology in the field of embedded system, computer and network security, and intelligence system as well as innovative and productive ideas related to emerging technology and computer engineering, including but not limited to : Network Security System Security Information Security Social Network & Digital Security Cyber Crime Machine Learning Decision Support System Intelligent System Fuzzy System Evolutionary Computating Internet of Thing Micro & Nano Technology Sensor Network Renewable Energy Wearable Devices Embedded Robotics Microcontroller
Arjuna Subject : Ilmu Komputer (semua kategori) - Kecerdasan Buatan
Articles 240 Documents
 20   21   22   23   24 
Automated Assessment of Research Grant Proposals Using Hybrid Semantic-Tabular Machine Learning: An Application to the SRIKANDI Research Management System Edwin Hari Agus Prastyo; Meriana Wahyu Nugroho; Reza Augusta Jannatul Firdaus; Tanhella Zein Vitadiar
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 2 (2026): June 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i2.12634

Abstract

Purpose – This study aims to develop and evaluate a hybrid semantic-tabular machine learning framework for supporting the automated assessment of research grant proposals in the SRIKANDI Research Management System at Universitas Hasyim Asy’ari Tebuireng Jombang. Design/methods/approach – The study employed a computational experimental design using historical institutional proposal data from the SRIKANDI system. A total of 190 proposals were labeled based on the institutional LPPM scoring threshold, consisting of 107 approved and 83 not approved proposals. The proposed framework integrates semantic features extracted from proposal narratives using IndoBERT with structured tabular features, including document completeness, proposal score, text quality, budget information, research field, and proposer track-record indicators. The fused 786-dimensional feature representation was classified using a Random Forest model with balanced class weights. Model performance was evaluated using accuracy, precision, recall, F1-score, ROC-AUC, confusion matrix analysis, and SHAP-based explainability. Findings – The hybrid model achieved strong predictive performance on the holdout test set, with 96.55% accuracy, 96.54% weighted F1-score, 96.75% weighted precision, 96.55% weighted recall, and ROC-AUC of 1.0000. The model correctly classified all approved proposals and misclassified only one not-approved proposal. SHAP analysis showed that word count, reference count, proposal text quality, and proposal score were the most interpretable contributors, while IndoBERT semantic dimensions added meaningful predictive value beyond administrative features. Research implications/limitations – The findings indicate that hybrid semantic-tabular learning can support more consistent and transparent preliminary proposal screening. However, the study is limited by its single-institution dataset, relatively small sample size, high feature-to-sample ratio, and dependence on score-threshold labeling. Originality/value – This study contributes a replicable explainable AI framework that combines Indonesian-language semantic representation, institutional tabular features, and SHAP-based interpretability for research grant proposal assessment within a university research management system.
Simple Additive Weighting Method in Determining Tourism Destination Recommendations at Onrust Archaeological Park Fata Nidaul Khasanah; Sugeng Murdowo; Ari Pambudi
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 2 (2026): June 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i2.12803

Abstract

Purpose – This study aims to apply the Simple Additive Weighting (SAW) method to recommend tourism destinations within Onrust Archaeological Park, Kepulauan Seribu, based on attraction, accessibility, and amenity criteria. Design/methods/approach – Primary data were collected through a questionnaire administered to 45 tourists who had visited Kelor Island, Onrust Island, or Cipir Island. Respondents were selected using convenience sampling and evaluated five criteria, namely scenery, photo spots, transportation, toilets, and dining facilities, using a five-point Likert scale. The average criterion scores were normalized to determine proportional weights. The SAW method was then employed to normalize the decision matrix, calculate individual preference values, aggregate the scores, and rank the three tourism destinations. Findings – Toilet availability received the highest criterion weight of 0.406, followed by photo spots at 0.281, scenery at 0.188, transportation at 0.063, and dining facilities at 0.063. The aggregated results show that Kelor Island achieved the highest average preference value of 0.868, followed by Onrust Island at 0.813 and Cipir Island at 0.777. Therefore, Kelor Island was identified as the most recommended destination based on the criteria evaluated. Research implications/limitations – The findings provide practical information for tourism managers and policymakers in identifying destination development priorities, particularly improvements in basic visitor facilities. However, the use of convenience sampling, the relatively small sample of 45 respondents, and destination-specific evaluations limit the generalizability of the findings. Originality/value – This study extends the application of the SAW method by integrating attraction, accessibility, and amenity dimensions to evaluate historical island destinations within the specific context of Onrust Archaeological Park.
UAV Based Automated Surveillance of Ganoderma boninense in Oil Palm Canopies Using YOLO26 Architecture Muhammad Rizky Pribadi; Hafiz Irsyad; Eka Puji Widiyanto; Muhammad Tri Setianto; Safeti Intan Pratiwi
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.11502

Abstract

Purpose – This study develops and evaluates a UAV-based automated surveillance approach using the YOLO26 architecture to detect visible symptoms associated with Ganoderma boninense infection in oil palm canopies. The study addresses the limitations of conventional manual inspection and multi-stage detection systems by applying a unified single-stage object-detection framework. Design/methods/approach – The model was developed using a publicly available dataset containing 1,133 annotated UAV images of oil palm canopies. Images were preprocessed, augmented, and partitioned using a plantation-block-aware strategy to reduce spatial data leakage. YOLO26s was trained using the Ultralytics framework on an NVIDIA Tesla T4 GPU. Model performance was evaluated using precision, recall, mAP@50, mAP@50–95, confidence-threshold sensitivity analysis, precision–recall curves, and five-fold block-aware cross-validation. Findings – On the independent block-aware test set of 118 images, the model achieved an mAP@50 of 77.14%, mAP@50–95 of 41.98%, precision of 66.03%, and recall of 76.32%. Five-fold block-aware cross-validation produced a mean mAP@50 of 74.12% ± 5.51% and a mean mAP@50–95 of 41.31% ± 4.00%. The relatively high recall indicates that the model can identify most visible infection instances, although its moderate precision shows that false-positive detections remain a practical concern. Research implications/limitations – The findings demonstrate the potential of YOLO26 to support UAV-based oil palm disease surveillance and targeted field inspection. However, the study relies on a single public dataset with inherited annotation procedures, lacks geographically independent external validation, and does not include direct benchmarking on onboard UAV or embedded edge devices. Originality/value – This study provides an early empirical evaluation of YOLO26 for UAV-based detection of visible Ganoderma symptoms in oil palm canopies. Its contribution lies in combining a single-stage detection architecture with plantation-block-aware evaluation, threshold-sensitivity analysis, and cross-validation to provide a more leakage-controlled assessment of model performance.
Sensitivity Analysis of Weight Normalization Schemes in McCall-Based Software Quality Measurement Ahmad Farisi; Dafid
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.11711

Abstract

Purpose – This study examines the sensitivity of McCall-based software quality measurement to alternative weight normalization intervals and rounding configurations. It specifically evaluates whether different weighting schemes produce consistent Total Quality scores and quality classifications when the evaluated system, criteria values, and aggregation procedures remain unchanged. Design/methods/approach – A quantitative experimental design was applied to the Product Operation dimension of the McCall model using Simponi, a web-based learning management system at Universitas Multi Data Palembang, as the case study. Criteria values were obtained from 166 student respondents, while indicator importance ratings were provided by two software engineering experts. Three normalization intervals, namely 0.1–0.5, 0.1–0.4, and 0.4–0.8, were evaluated using floor and ceiling rounding configurations. The resulting scores were calculated through hierarchical weighted additive aggregation. Findings – Weight normalization substantially influenced the resulting software quality assessment. The moderate and expanded weighting schemes produced comparable Total Quality outcomes and consistently classified the evaluated system at the highest quality level, whereas the compressed weighting scheme generated a lower overall assessment and shifted the system into a lower quality category. In contrast, differences between floor and ceiling rounding configurations were negligible and did not alter the resulting quality classification. Research implications/limitations – Weight normalization should be explicitly documented and standardized to improve comparability across McCall-based evaluations. The findings are limited to the Product Operation dimension, three predefined intervals, and one academic web application. Originality/value – This study provides empirical sensitivity evidence showing that weight interval selection is a substantive methodological decision that can alter software quality interpretations, whereas rounding configurations have negligible practical effects.
Edge-Based Early Warning for High-Speed Boat Stability Monitoring Muhammad Asep Subandri; Jamal; Fajar Ratnawati; I Gusti Agung Putu Mahendra; Agus Tedyyana; Budhi Santoso
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.11866

Abstract

Purpose - This study aims to develop and evaluate an edge-based early warning prototype for monitoring the stability of high-speed boats using real-time motion data. Design/methods/approach – The study employs an engineering prototype validation design consisting of system requirement analysis, architecture design, implementation, and validation. The system integrates an IMU (MPU-6050) for motion sensing, a Raspberry Pi-based edge computing unit for real-time processing, rule-based classification (Normal/Warning/Critical), and an MQTT-based communication framework connected to the SHISTAMO dashboard. Prototype validation includes functional testing, platform integration, and operational monitoring using controlled scenarios and expert-labeled events. Findings - The results show that the prototype successfully performs end-to-end integration from sensing to visualization. The system achieved 90.4% classification accuracy, with high recall in detecting critical conditions (96.7%), ensuring reliable identification of high-risk events. The local alarm response time was 182 ms, while the dashboard update delay averaged 1.24 s, indicating near-real-time performance. Communication reliability was also high, with 98.8% data delivery success and 97.2% offline synchronization. Research implications/limitations – The findings demonstrate prototype-level feasibility; however, validation is limited to controlled scenarios and does not yet represent diverse sea conditions. The rule-based thresholds and comfort proxy require further calibration and validation through extended sea trials and reference instrumentation. Originality/value – This study contributes an integrated edge-based maritime monitoring prototype that combines motion sensing, offline-capable alarming, real-time telemetry, and fleet-level logging in a single system, specifically tailored to the operational needs of high-speed boats.
Faculty Academic Web Security Assessment via Grey-Box VAPT and CVSS v3.1 Akbar; Muh.Riyaldi Pratama; Akbar Iskandar; Riska Khaerani; Kamaruddin; Listia Utami
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.11921

Abstract

Purpose – This study evaluates the security posture of a faculty-level academic web application by applying a grey-box Vulnerability Assessment and Penetration Testing (VAPT) approach and classifying validated vulnerabilities using the Common Vulnerability Scoring System (CVSS) v3.1. Design/methods/approach – An evaluative case study was conducted on a live Research Registration Information System using an authenticated non-administrative account. The assessment combined attack-surface mapping, automated vulnerability scanning, HTTP request–response observation, controlled request manipulation, and repeated manual validation. Each suspected vulnerability was evaluated through at least three controlled request variations and was classified as confirmed only when its behavior was reproducible, security-relevant, and distinguishable from normal application behavior or false-positive detection. Confirmed vulnerabilities were subsequently assessed using CVSS v3.1. Findings – Automated and manual assessment produced 14 candidate findings, of which three (21.4%) were confirmed after analytical validation and 11 were rejected as non-reproducible or false positives. The validated vulnerabilities comprised SQL injection, authentication bypass, and cross-site scripting (XSS) associated with file upload functionality. All three were classified as high severity, with CVSS v3.1 base scores of 8.8, 8.3, and 7.6, respectively. The findings indicate weaknesses across backend input processing, authentication and session control, and user-generated content handling, suggesting that security risks extend across multiple operational layers of the application. Research implications/limitations – The results demonstrate the importance of combining automated detection with manual validation to improve the reliability of web security assessments and support risk-based remediation. However, the study is limited to a single faculty-level system and a specific grey-box access context, which restricts direct generalization to other institutional architectures. Originality/value – Rather than proposing a new security framework, this study provides a transparent and traceable application of grey-box VAPT at the underreported faculty-subdomain level, linking reproducible technical findings with CVSS-based severity prioritization and practical cybersecurity governance implications.
Demographic Segmentation of Election Supervisors Using K-Means Clustering Kirei D.T Palar; Irene R.H.T. Tangkawarouw; Sondy C. Kumajas
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.12205

Abstract

Purpose – This study demonstrates the use of K-Means clustering for demographic segmentation of Indonesian election supervisors and proposes a dashboard-based analytical prototype for workforce profiling. The study responds to the absence of systematic, data-driven segmentation in supervisor development, training, mentoring, and resource allocation. Design/methods/approach – A quantitative data science workflow was applied, covering synthetic data generation, preprocessing, clustering, validation, stability testing, and dashboard development. Because formal requests for real supervisor-level demographic data from Bawaslu were denied due to privacy, consent, and re-identification concerns, the study used a synthetic dataset of 500 supervisors generated from publicly available institutional and demographic parameters. Four variables were used as clustering inputs: age, years of experience, gender, and education level. K-Means clustering was implemented with standardized features, and the optimal number of clusters was evaluated using the Elbow Method and Silhouette Score. Findings – The analysis identified three illustrative segments: junior-like, mid-level-like, and senior-like supervisor profiles. The optimal cluster solution was K=3, supported by the Elbow Method and a Silhouette Score of 0.38, indicating moderately well-defined clusters. Stability testing showed consistent results across multiple random seeds. The Streamlit dashboard successfully visualized demographic distributions, cluster profiles, and provincial dominance patterns. Research implications/limitations – The findings provide a methodological prototype only and should not be interpreted as empirical evidence about Bawaslu’s actual workforce. Originality/value – The study contributes by applying clustering to election supervisor segmentation, integrating geospatial dashboard visualization, and transparently documenting synthetic-data use when real administrative data access is restricted in a sensitive institutional context.
Accuracy–Efficiency Trade-off Analysis of Five Lightweight CNN Architectures for Mobile-Deployable Corn Leaf Disease Classification Jarot Budiasto; Hasanudin Jayawardana; Tri Kustanti Rahayu; Tatik Melinda Tallulembang
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.12264

Abstract

Purpose – Corn leaf disease diagnosis in resource-constrained agricultural settings requires mobile-deployable models that maintain a practical balance between classification accuracy, model size, and on-device latency. This study aims to provide empirical guidance for selecting lightweight Convolutional Neural Network (CNN) architectures by systematically analyzing the accuracy–efficiency trade-offs of five models for corn leaf disease classification. Design/methods/approach – MobileNetV2, MobileNetV3-Small, MobileNetV3-Large, EfficientNetB0, and NASNetMobile were evaluated on the PlantVillage Corn dataset comprising 4,188 images across four classes under identical experimental settings. The models were trained using a two-phase strategy and converted into standard and dynamic-range quantized TensorFlow Lite formats. Evaluation covered classification accuracy, macro F1-score, model size, Android on-device inference latency, Pareto frontier and radar analyses, and pairwise McNemar's tests with Yates continuity correction. Findings – EfficientNetB0 achieved the highest accuracy (95.25%) and macro F1-score (93.77%). MobileNetV3-Small offered the strongest efficiency under the tested Android CPU setting, reaching 94.54% accuracy with a 1.18 MB dynamic-range quantized TensorFlow Lite model and 3.89 ± 0.04 ms standard inference. The top three models were statistically comparable (p = 0.6625-1.0000). Research implications/limitations – Standard TensorFlow Lite is preferable for low-latency Android CPU deployment, whereas dynamic-range quantized TensorFlow Lite supports storage-constrained offline distribution. However, the findings are limited to the PlantVillage benchmark dataset and testing on a single mid-range Android device. Originality/value – This study integrates lightweight CNN benchmarking, TensorFlow Lite deployment, real-device Android testing, accuracy–efficiency trade-off analysis, and statistical validation to support evidence-based mobile agricultural AI model selection.
Design and Implementation of an IoT-Based Multi-Parameter Radiator Coolant Quality Monitoring System for Predictive Maintenance Habib Roviurrahman; Buang Turasno; Dzaki Putra Prakosa; Ramadhan Dwi Prasetyo
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.12799

Abstract

Purpose – This study develops and evaluates an Internet of Things (IoT)-based multi-parameter monitoring system for assessing radiator coolant quality and supporting predictive maintenance. The system continuously monitors coolant volume, pH, temperature, and turbidity while providing real-time status information and anomaly alerts through a mobile application. Design/methods/approach – The study employed a Research and Development approach involving system design, prototype development, sensor calibration, laboratory validation, and field implementation. The prototype integrated an ESP32 microcontroller with ultrasonic, pH, temperature, and optical sensors. Sensor data were transmitted through Firebase Realtime Database and displayed using a mobile application developed with MIT App Inventor. Validation was performed against calibrated reference instruments, followed by field testing under urban driving conditions and usability evaluation using the System Usability Scale. Findings – The proposed system demonstrated high measurement accuracy across all monitored coolant parameters and maintained reliable real-time communication between the embedded device and mobile application. Field testing showed that the system successfully identified coolant leakage, acidification, and excessive turbidity before noticeable vehicle-performance degradation occurred. The mobile application also demonstrated favorable usability, indicating that the monitoring interface was practical and accessible for users. Research implications/limitations – The proposed system demonstrates potential for supporting real-time coolant-condition monitoring and early maintenance intervention. However, validation was limited to one vehicle, a 14-day testing period, and urban driving conditions. Originality/value – This study integrates four coolant-quality parameters within a unified IoT architecture, offering a more comprehensive predictive-maintenance approach than conventional single-parameter radiator monitoring systems.
Internet of Things-Based Decision Support System for Toddler Health Using Mamdani Fuzzy Logic I Gede Wiryawan; Yogiswara; Beni Widiawan; Lalitya Nindita Sahenda; Nanda Raditya Akbar
Journal of Embedded Systems, Security and Intelligent Systems Vol 7 No 3 (2026): September 2026
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/jessi.v7i3.13086

Abstract

Purpose – This study develops and evaluates an Internet of Things-based decision support system using Mamdani fuzzy logic for the early identification of febrile seizure risk in toddlers. The system integrates continuous body-temperature and heart-rate monitoring to provide automated risk classification and real-time warning information. Design/methods/approach – The prototype combines an Arduino Uno, a GY-906 non-contact temperature sensor, a pulse sensor, an ESP8266 communication module, and a web-based monitoring platform. Body temperature and heart rate were processed through a Mamdani fuzzy inference system comprising fuzzification, nine clinical IF–THEN rules, MAX rule aggregation, and centroid defuzzification. Hardware performance was validated against standard clinical instruments using 30 toddler measurements. System classification was compared with risk assessments independently assigned by a senior pediatrician, while end-to-end latency was evaluated through 30 transmission cycles. Findings – The GY-906 sensor achieved 98.92% accuracy with a mean squared error of 0.214, while the pulse sensor achieved 96.35% accuracy with a mean squared error of 12.67. The fuzzy inference system correctly classified 29 of 30 cases, resulting in 96.67% accuracy, 100% sensitivity, and 93.33% specificity. The system also achieved an average end-to-end latency of 1.877 seconds, indicating responsive real-time monitoring under controlled laboratory network conditions. Research implications/limitations – The proposed system may support caregivers and healthcare professionals by providing consistent and timely early-warning information. However, the prototype was validated using a relatively small sample, depends on stable network connectivity, and requires broader prospective clinical validation before routine medical implementation. Originality/value – This study integrates real-time multivariable IoT monitoring with Mamdani fuzzy reasoning to transform uncertain pediatric vital-sign boundaries into interpretable febrile seizure risk classifications.

Page 23 of 24 | Total Record : 240

 20   21   22   23   24 

Filter by Year

2020 2026


Reset
Filter By Issues
All Issue Vol 7 No 3 (2026): September 2026 Vol 7 No 2 (2026): June 2026 Vol 7 No 1 (2026): March 2026 Vol 6, No 4 (2025): Desember 2025 Vol 6, No 3 (2025): September 2025 Vol 6, No 2 (2025): June 2025 Vol 6, No 1 (2025): March 2025 Vol 5, No 3 (2024): November 2024 Vol 5, No 2 (2024): July 2024 Vol 5, No 1 (2024): March 2024 Vol 4, No 2 (2023): November 2023 Vol 4, No 1 (2023): May 2023 Vol 3, No 2 (2022): November 2022 Vol 3, No 1 (2022): May 2022 Vol 2, No 2 (2021): November 2021 Vol 2, No 1 (2021): May 2021 Vol 1, No 2 (2020): November 2020 Vol 1, No 1 (2020): May 2020 More Issue