cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Abd. Charis Fauzan
Contact Email
fauzancharis@gmail.com
Phone
+6287750503014
Journal Mail Official
-
Editorial Address
Jl. Masjid Nomor 22 Kota Blitar, Jawa Timur
Location
Kab. blitar,
Jawa timur
INDONESIA
ILKOMNIKA: Journal of Computer Science and Applied Informatics
Published by Universitas Nahdlatul Ulama Blitar
ISSN : -     EISSN : 27152731     DOI : https://doi.org/10.28926/ilkomnika
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management
ILKOMNIKA: Journal of Computer and Applied Informatics is is a peer reviewed open-access journal. The journal invites scientists and engineers throughout the world to exchange and disseminate theoretical and practice-oriented topics of computer science and applied informatics which covers five (5) majors areas of research that includes 1) Informatics Engineering and Its Application 2) Computer Science 3) Software Engineering 4) Computer Engineering 5) Information System. This journal is published 3 issues a year, in April, August, and December.
Arjuna Subject : Ilmu Komputer (semua kategori) - Aplikasi Ilmu Komputer
Articles 229 Documents
 19   20   21   22   23 
Generalization Analysis of a Long Short-Term Memory Model for Cross-Domain Malware Detection Prasetyo, Stefanus Eko; Haeruddin, Haeruddin; Jason, Jason
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.856

Abstract

The increasing diversity of malware targeting heterogeneous computing environments poses significant challenges to conventional detection approaches that rely on domain-specific assumptions. In particular, detection models optimized for a single dataset often exhibit limited robustness when applied to data with different structural and behavioral characteristics. This study analyzes the generalization capability of a Long Short-Term Memory (LSTM) model for behavior-based malware detection across multiple domains. A fixed two-layer LSTM architecture is evaluated using one primary dataset, CIC-MalMem-2022, and four additional datasets representing Android applications, Internet of Things network traffic, botnet behavior, and static Windows Portable Executable analysis. Although each dataset undergoes a dataset-specific preprocessing pipeline, all experiments employ an identical model architecture and hyperparameter configuration to ensure consistent and comparable evaluation. Model performance is assessed using standard classification metrics, supported by single train–test evaluation and five-fold cross-validation to examine performance stability and robustness. The experimental results demonstrate that the LSTM model maintains consistently high detection performance across datasets with diverse characteristics, including both sequential and non-sequential data representations. These findings indicate that the model effectively captures fundamental malware behavior patterns that generalize beyond a single domain, highlighting its potential applicability in heterogeneous cybersecurity environments where cross-domain robustness is required. At the same time, the evaluation is conducted under controlled experimental conditions and does not explicitly address adversarial adaptation or fully dynamic runtime deployment, which should be considered when interpreting the results for practical operational use.
Lifestyle-Based Obesity Risk Clustering Using Ward Hierarchical Clustering Aunilla, Moch. Fikri; Mufliq, Achmad; Nugroho, Rizky Aditya
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.860

Abstract

Obesity is a growing public health problem influenced by multiple interacting lifestyle behaviors that cannot always be adequately captured by single-factor or label-driven analysis. Therefore, this study applies an unsupervised clustering approach to identify natural behavioral segments associated with obesity risk. The data were obtained from the Poltekkes Kemenkes Semarang Obesity Risk Dataset, consisting of 20,758 records and 16 mixed attrib`utes (numerical and categorical) with the NObeyesdad label. Pre-processing included standardizing numerical features and one-hot encoding categorical features, followed by dimensionality reduction using PCA to 13 components retaining approximately 95.48% of the variance. Ward clustering was applied in the PCA space, and the number of clusters was tested for k=2–10 using the Silhouette coefficient, Davies–Bouldin Index (DBI), and Calinski–Harabasz (CH) index. Although the average Silhouette coefficient was modest (≈0.2029), the k=5 solution was retained because it offered the best balance between internal validation results and the practical interpretability of cluster-based risk profiles. BMI-based interpretation using WHO Asia criteria identified Cluster 0 as very high risk (mean BMI 33.04; 73.7% obese), Cluster 2 as high risk characterized by predominant smoking (65.3% obese), Cluster 4 as moderate-to-high risk (34.0% overweight; 30.3% obese), Cluster 1 as a mixed group, and Cluster 3 as relatively low risk (mean BMI 22.21; 8.9% obese). Agreement between clusters and the label was low (NMI 0.126; ARI 0.073), indicating that the clusters represent similarity in behavioral patterns rather than the label classes.
Fine-Tuned BART Transfer Learning for Abstractive Summarization of Indonesian YouTube Transcripts with ROUGE Evaluation Setiawan, Diyan Nova; Faisal, Muhammad; Imamudin, Mochamad
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.862

Abstract

Indonesian YouTube transcripts present substantial challenges for abstractive summarization because they contain informal expressions, filler words, fragmented utterances, and automatically generated caption errors. Existing Indonesian summarization studies have mostly focused on formal written texts such as news articles, while duration-aware abstractive summarization of noisy Indonesian educational video transcripts remains underexplored. This study fine-tunes a BART-based sequence-to-sequence model using a curated corpus of Indonesian YouTube transcripts from the “Kok Bisa?” educational channel. From 1,000 collected videos, 957 transcripts were successfully retrieved, and 730 transcripts passed the final filtering criteria for experimental analysis. The dataset was divided into short, medium, and long transcript categories to evaluate the effect of input duration on summarization quality. The proposed pipeline includes transcript retrieval, metadata extraction, text normalization, filler-word removal, repetition filtering, BART fine-tuning, summary generation, and ROUGE-based evaluation. The model achieved the best performance on short transcripts, with ROUGE-1 F1 = 0.621, ROUGE-2 F1 = 0.438, and ROUGE-L F1 = 0.587. Performance decreased on long transcripts, with ROUGE-1 F1 = 0.552, ROUGE-2 F1 = 0.384, and ROUGE-L F1 = 0.509, indicating that longer narratives reduce lexical and structural alignment. These findings show that fine-tuned BART is effective for short Indonesian educational transcripts but requires segmentation, semantic evaluation, and stronger baseline comparison for long-form video summarization.
Early Detection of Phishing, Disinformation, and Extreme Opinions in Digital Text Using Transformer-Based Models Wibowo, Munif; Faisal, Muhammad; Nugroho, Fresy
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.863

Abstract

The rapid expansion of digital communication platforms has increased the circulation of phishing messages, disinformation, and extreme opinions, creating urgent challenges for cybersecurity and social stability. This study proposes a hybrid CNN–BiLSTM–Transformer framework for the early detection of harmful digital text. The model integrates convolutional feature extraction, sequential dependency learning, and self-attention mechanisms to capture local lexical patterns, contextual relations, and long-range semantic dependencies. Experimental evaluation was conducted using accuracy, precision, recall, F1-score, and ROC analysis, with CNN, LSTM, and RoBERTa used as baseline models. The proposed hybrid model achieved the highest classification accuracy of 95.0%, outperforming CNN (86.0%), LSTM (88.0%), and RoBERTa (91.0%). In addition, the model obtained 90.0% precision, 93.0% recall, and 91.5% F1-score, indicating a balanced ability to reduce false positives while maintaining strong detection sensitivity. Robustness testing further showed that the F1-score remained stable across normal, noisy, and adversarial text conditions, decreasing from 95.0% under normal conditions to 92.0% and 90.0% under noisy and adversarial settings, respectively. These findings demonstrate that the proposed hybrid Transformer-based architecture provides an effective and robust approach for supporting automated Cyber Early Warning Systems in detecting harmful digital content.
GA-Optimized Stacking Ensemble for Unified Phishing Website and Email Detection Using Multi-Domain Feature Representation Hermawan, Hendra; Faisal, Muhammad; Imamudin, Mochamad
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.864

Abstract

Phishing remains a major cybersecurity threat because attacks increasingly combine fraudulent websites with deceptive email content. Existing detection models often focus on a single domain, such as URLs or emails, which limits their ability to capture heterogeneous phishing patterns. This study proposes a GA-optimized stacking ensemble framework for unified phishing website and email detection using multi-domain features. The framework combines URL structural attributes, email metadata, and semantic content features, while a Genetic Algorithm is used to reduce feature redundancy and select the most informative attributes. The proposed model is evaluated against baseline Random Forest, Gradient Boosting, and conventional Stacking classifiers using Accuracy, F1-score, AUC-ROC, cross-validation stability, robustness under noise, and inference latency. Experimental results show that the proposed GA-Stacking model achieves 98.1% accuracy, 97.6% F1-score, and 0.947 AUC-ROC, outperforming Random Forest, Gradient Boosting, and standard Stacking models. The model also reduces the feature set from 72 to 31 features and maintains strong robustness under simulated noise, with F1-score remaining at 92.0% under 30% perturbation. These findings indicate that evolutionary feature optimization improves the stability, efficiency, and robustness of stacking ensemble learning for multi-domain phishing detection.
A Systematic Review of AIoT in Agricultural Environmental Monitoring: A Comparative Analysis of Machine Learning Approaches Kharisma, Derry Artha; Sunanto, Sunanto
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.865

Abstract

Agricultural water management increasingly deploys AIoT systems that integrate IoT sensing with machine learning, yet deployable autonomous irrigation control remains largely unrealized despite widely reported accuracy above 90%. This systematic literature review of 25 peer-reviewed studies (2020–2026), conducted across five databases following PRISMA 2020 guidelines, diagnoses why predictive performance fails to translate into operational autonomy. The analysis identifies six interdependent structural gaps: open-loop prediction architectures, informationally narrow sensing, correlated co-sensor packaging, static non-adaptive models, accuracy–deployability decoupling, and metric inconsistency. These gaps form a dependency chain across data, inference, and actuation layers in which closed-loop integration depends on resolving data adequacy. A parallel finding reveals systematic methodological divergence between national and international research contexts, driven by infrastructure and deployment constraints rather than research quality, with reinforcement learning, hybrid multi-modal architectures, and continual learning largely absent in national studies. This study contributes a reframing of AIoT system maturity by demonstrating that within-study accuracy is misaligned with operational validity. It further establishes environmental generalizability as a more appropriate evaluation criterion, shows that the six structural gaps form a sequential dependency structure that prevents single-gap solutions from producing deployable improvement, and provides directional evidence that reported accuracy and validation scope are inversely related across the corpus, suggesting that current performance claims systematically overstate operational readiness.
Development of an IoT-Based Smart Scarecrow System with Multi-Sensor Integration for Bird Pest Control in Rice Fields Ramadhanty, Tarisha; Agustina, Dian Tri; Khusnul Khotimah, Triana Selvia; Ekawati, Ekawati; Zain, Muhammad Yasir; Rachmatullah, Sholeh
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.870

Abstract

This study aims to develop an IoT-based Smart Scarecrow system, an automated bird repellent device that integrates PIR sensors, ultrasonic sensors, an ESP32-CAM camera, LED flashers, ultrasonic buzzers, and a 360° rotation mechanism using a stepper motor. The system is controlled by an ESP32 microcontroller and powered by a solar panel as an independent energy source. The research methodology includes a literature review, system design, hardware and software implementation, laboratory testing, and limited environmental simulation to evaluate system performance. The evaluation focuses on response time, detection accuracy, and operational stability. The results indicate that the system is capable of responding to detected objects in less than 3 seconds, with a real-time visual transmission delay of under 2 seconds. The integration of multiple sensors enhances detection accuracy and reduces false positives. Compared to conventional scarecrow systems, the proposed system demonstrates greater adaptability due to its dynamic response and wider coverage area. However, this study is limited to prototype-scale testing, and further large-scale field validation is required. Overall, the system shows strong potential as an adaptive technological solution for supporting precision agriculture.
Optimizing MobileNetV2 Using Transfer Learning and Fine-Tuning Techniques for Lung Cancer Classification Rozi, Atiqur; Puspaningrum, Eva Yulia; mandyartha, Eka Prakarsa
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.871

Abstract

Lung cancer remains one of the leading causes of mortality worldwide, highlighting the importance of early and accurate detection. This study proposes a deep learning-based approach for lung cancer classification using the MobileNetV2 architecture on CT-scan images. Two experimental scenarios were investigated: transfer learning with a frozen base model and fine-tuning by unfreezing selected layers. The dataset was compiled from publicly available sources and balanced to address class imbalance. The model was trained using the Stochastic Gradient Descent (SGD) optimizer and evaluated using accuracy, precision, recall, and F1-score. The results demonstrate that the fine-tuning strategy achieves superior performance across most evaluation metrics compared to transfer learning. In particular, recall shows a significant improvement, indicating enhanced capability in detecting positive cancer cases, although accompanied by a slight decrease in precision. The F1-score also improves, reflecting a better balance between precision and recall. These findings suggest that fine-tuning enhances feature representation and improves classification performance within the experimental setting. However, the results are limited to the dataset used in this study, and further validation on larger and clinically representative datasets is required before considering real-world medical applications.
Integration of WhatsApp Business API and Artificial Intelligence in a Assisted-Automated Letter Generation System for Village Governance Priambodo, Gumilang; Dewi, Renny Sari; Candra, Ika Diyah; Dhenabayu, Riska
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.876

Abstract

Conventional rural administration in Indonesia faces manual clerical bottlenecks, while existing digital solutions often fail due to the digital divide and an inability to process unstructured local dialects. To address these challenges, this study designed and evaluated an Intelligent Robotic Process Automation (IRPA) prototype for fully autonomous village correspondence, specifically handling seven distinct types of official letters. Utilizing the Design Science Research Methodology (DSRM), the proposed architecture integrates the WhatsApp Business API as an inclusive conversational interface, Google Gemini for cognitive intent classification, and the n8n low-code platform for cloud-based document orchestration. Functional evaluation using 20 test prompts, which represent real-world informal language, abbreviations, and typographical errors, demonstrated that the cognitive agent achieved 100% accuracy in intent recognition and boundary detection. Furthermore, the system significantly reduced administrative Turnaround Time (TAT) by approximately 77.5%, effectively transforming manual processes of 15 to 25 minutes into a 3 to 6 minutes automated cycle. Ultimately, this research offers three main contributions: an asynchronous architecture to lower user cognitive load, a deterministic prompt engineering method for public services, and empirical evidence of RPA efficiency in inclusive rural governance.

Page 23 of 23 | Total Record : 229

 19   20   21   22   23 

Filter by Year

2019 2026


Reset
Filter By Issues
All Issue Vol 8 No 1 (2026): Volume 8, Number 1, April 2026 Vol 7 No 3 (2025): Volume 7, Number 3, December 2025 Vol 7 No 2 (2025): Volume 7, Number 2, August 2025 Vol 7 No 1 (2025): Volume 7, Number 1, April 2025 Vol 6 No 3 (2024): Volume 6, Nomor 3, Desember 2024 Vol 6 No 2 (2024): Volume 6, Nomor 2, Agustus 2024 Vol 6 No 1 (2024): Volume 6, Nomor 1, April 2024 Vol 5 No 3 (2023): Volume 5, Nomor 3, Desember 2023 Vol 5 No 2 (2023): Volume 5, Nomor 2, Agustus 2023 Vol 5 No 1 (2023): Volume 5, Nomor 1, April 2023 Vol 4 No 3 (2022): Volume 4, Nomor 3, Desember 2022 Vol 4 No 2 (2022): Volume 4, Nomor 2, Agustus 2022 Vol 4 No 1 (2022): Volume 4, Nomor 1, April 2022 Vol 3 No 3 (2021): Volume 3, Nomor 3, Desember 2021 Vol 3 No 2 (2021): Volume 3, Nomor 2, Agustus 2021 Vol 3 No 1 (2021): Volume 3, Nomor 1, April 2021 Vol 2 No 3 (2020): Volume 2, Nomor 3, Desember 2020 Vol 2 No 2 (2020): Volume 2, Nomor 2, Agustus 2020 Vol 2 No 1 (2020): Volume 2, Nomor 1, April 2020 Vol 1 No 2 (2019): Volume 1, Nomor 2, Desember 2019 Vol 1 No 1 (2019): Volume 1, Nomor 1, Agustus 2019 More Issue