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 
Pemantauan Keamanan Real-Time pada Base Transceiver Station (BTS) Menggunakan YOLOv8 dan Integrasi Telegram: Optimasi Model dan Evaluasi Performa Sobirin, Muhammad; Wijonarko, Panji
ILKOMNIKA Vol 7 No 3 (2025): Volume 7, Number 3, December 2025
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v7i3.824

Abstract

Infrastruktur Base Transceiver Station (BTS) memiliki peran vital dalam telekomunikasi namun menghadapi risiko keamanan fisik yang tinggi, seperti pencurian dan vandalisme, yang dapat mengganggu ketersediaan jaringan. Metode pemantauan manual yang ada saat ini dinilai tidak efisien karena membutuhkan sumber daya intensif dan memiliki latensi respon yang tinggi terhadap pelanggaran keamanan. Penelitian ini mengusulkan optimalisasi sistem pengawasan otomatis real-time menggunakan algoritma Deep Learning YOLOv8 yang terintegrasi dengan notifikasi IoT berbasis Telegram. Empat varian arsitektur model (YOLOv8n, YOLOv8n-p2, YOLOv8n-p6, dan YOLOv8s) dievaluasi secara komparatif menggunakan metrik mean Average Precision (mAP), F1-score, dan kecepatan inferensi (Frames Per Second/FPS). Evaluasi dilakukan menggunakan Human Dataset yang terdiri dari 17.300 citra dengan pelatihan pada platform Google Colab dan pengujian pada perangkat edge NVIDIA Jetson Nano. Hasil eksperimen menunjukkan adanya trade-off signifikan antara akurasi dan kecepatan; YOLOv8s mencapai akurasi tertinggi dengan mAP@0.5 sebesar 61,4%, namun dengan kecepatan inferensi rendah (9,67 FPS). Sebaliknya, YOLOv8n menawarkan keseimbangan optimal dengan mAP@0.5 sebesar 59,3% dan kecepatan tertinggi mencapai 22,02 FPS. Sementara itu, varian modifikasi YOLOv8n-p2 (14,84 FPS) dan YOLOv8n-p6 (21,18 FPS) menunjukkan kemampuan kompetitif dalam menangani variasi skala objek namun tidak melampaui efisiensi YOLOv8n. Secara praktis, penelitian ini merekomendasikan implementasi YOLOv8n pada perangkat edge berdaya rendah karena kemampuannya memproses video secara real-time dan mengirimkan peringatan dini via Telegram secara instan, sehingga secara signifikan meningkatkan responsivitas sistem keamanan BTS.
Deep Learning‑Based Sentiment Classification on Category Service and Resolution of Consumer Complaints in Digital Banking Kusuma, Irnayanti Dwi; Fatichah, Chastine
ILKOMNIKA Vol 7 No 3 (2025): Volume 7, Number 3, December 2025
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v7i3.825

Abstract

The growth of digital banking in Indonesia has transformed customer interactions. It emphasizes the need to understand user sentiment and feedback. This study aims to analyze public perceptions of the Jenius digital banking application through sentiment analysis using deep learning methods enhanced by easy data augmented (EDA). The dataset written in Indonesian related to Jenius from Twitter. Data collected between August 2016 and August 2024 were manually annotated for sentiment polarity (positif, netral, negatif) and complaint handling categories (edukasi, konsultasi, fasilitasi, none). The EDA technique was used to enhance linguistic diversity and reduce class imbalance before training two deep learning models, Bidirectional Long Short-Term Memory (BiLSTM) and Convolutional Neural Network (CNN). The results show that EDA + BiLSTM achieved an accuracy of 0.68, whereas EDA + CNN obtained 0.66. BiLSTM slightly outperforms CNN across precision, recall, and F1-score. These findings indicate that both models effectively handle augmented data, with the BiLSTM model demonstrating a better contextual understanding of Bahasa Indonesia. The integration of EDA significantly improves the robustness and performance of the model in sentiment and aspect-based classification. This study highlights the potential of EDA as a simple yet effective method for enhancing deep learning models.
Sistem Rekomendasi Destinasi Wisata Menggunakan Data Demografis Berbasis Klasifikasi Support Vector Machine A'yun, Aldilla Qurrata; Arif, Yunifa Miftachul; ., Muhammad Imamudin
ILKOMNIKA Vol 7 No 3 (2025): Volume 7, Number 3, December 2025
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v7i3.826

Abstract

Penelitian ini bertujuan untuk mengembangkan sistem rekomendasi destinasi wisata berbasis data demografis menggunakan metode klasifikasi Support Vector Machine (SVM). Sistem dirancang untuk memberikan rekomendasi destinasi yang sesuai dengan karakteristik wisatawan, seperti usia, jenis kelamin, dan status sosial. Dataset yang digunakan terdiri dari sepuluh variabel demografis dan empat belas kategori destinasi wisata. Analisis awal menunjukkan bahwa dataset memiliki ketidakseimbangan kelas yang sangat tinggi, dengan kelas Jatim Park 1 mendominasi lebih dari separuh data sementara banyak kelas lain hanya memiliki 1–6 sampel. Untuk mengurangi dampak ketidakseimbangan ini, dilakukan teknik oversampling pada data training. Model SVM kemudian dilatih menggunakan beberapa kombinasi parameter dan kernel, serta diuji menggunakan metrik akurasi, precision, recall, dan F1-score. Hasil eksperimen menunjukkan bahwa pada data training yang sudah diseimbangkan, performa model meningkat signifikan, ditunjukkan oleh nilai F1-macro pada cross-validation sebesar 0.84. Namun, ketika diuji pada data testing yang mencerminkan kondisi distribusi asli, performa model menurun, dengan akurasi sebesar 54% dan nilai F1-macro yang rendah pada sebagian besar kelas minoritas. Temuan ini menunjukkan bahwa meskipun SVM efektif pada data yang seimbang, performanya masih belum optimal pada dataset rekomendasi wisata yang sangat timpang. Penelitian ini merekomendasikan pengayaan data, terutama untuk kelas-kelas minoritas, serta eksplorasi metode penanganan ketidakseimbangan kelas lainnya pada penelitian lanjutan.
Implementasi Sistem Penjaminan Mutu Eksternal Berbasis Website Menggunakan Pendekatan Evolutionary Prototype Taqiyuddin, Muhammad Akmal Faris; Anugrah, Indra Gita; Witra, Widyasari Puspa Permata
ILKOMNIKA Vol 7 No 3 (2025): Volume 7, Number 3, December 2025
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v7i3.834

Abstract

Universitas memikul tanggung jawab penting dalam menjaga standar pendidikan melalui mekanisme penjaminan mutu yang efektif. Dalam proses akreditasi program studi, dokumen seperti Laporan Kinerja Program Studi (LKPS) dan Laporan Evaluasi Diri (LED) harus disusun secara sistematis. Namun, belum terdapat solusi Sistem Penjaminan Mutu Eksternal (SPME) yang iteratif dan adaptif untuk mendukung pengelolaan dokumen, pemantauan performa mutu, serta verifikasi pemenuhan kriteria akreditasi secara terintegrasi. Penelitian ini bertujuan membangun SPME berbasis web menggunakan pendekatan Evolutionary Prototype yang memungkinkan pengembangan sistem secara bertahap berdasarkan umpan balik pengguna. Sistem diuji menggunakan Black Box Testing dengan 31 skenario uji yang melibatkan beberapa stakeholder (Dekan, Ketua Program Studi, DPM (Asesor)) untuk memvalidasi fungsionalitas sistem. Hasil pengujian menunjukkan bahwa semua skenario uji berjalan sesuai expected result, sistem menghasilkan laporan yang sistematis, dan memudahkan pengawasan performa mutu secara real-time. Prototype sistem ini berhasil mendemonstrasikan kemampuan dalam mendukung penjaminan mutu eksternal dan mempercepat proses pengumpulan serta pelaporan data akreditasi. Kontribusi praktis penelitian ini adalah pengurangan waktu dalam penyusunan LKPS karena data dapat diakses dan diolah secara digital, meningkatkan produktivitas dan akurasi pelaporan, serta meminimalkan kesalahan manual dalam pengolahan data akreditasi.
A Comparative Study of Extractive and Generative Approaches for Indonesian Meeting Minutes Summarization Harliana, Harliana; Sismoro, Heri
ILKOMNIKA Vol 7 No 3 (2025): Volume 7, Number 3, December 2025
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v7i3.846

Abstract

This study compares extractive and generative approaches for automatic summarization of Indonesian meeting minutes. Our main scientific contribution is an empirical claim that, under strict zero-shot conditions and without domain adaptation, simple extractive baselines are more reliable than off-the-shelf generative models in preserving both decision content and meeting-context cues (actors/roles). We evaluate three extractive baselines (Lead-3, Random-Extract, TextRank-Simple) against an Indonesian GPT-2 model tested under multiple decoding configurations and an mT5 sequence-to-sequence model in a zero-shot setting. Experiments utilize 30 manually curated meeting minutes. The dataset size is intentionally limited because meeting minutes are heterogeneous and require carefully constructed reference summaries to ensure evaluation validity; the study is positioned as a controlled diagnostic comparison rather than a training or adaptation effort. Performance is measured using ROUGE-1/2/L, summary–to–reference length ratios, simple audits of gender and professional role mentions, correlations between decoding parameters and ROUGE, and paired t-tests. Results show that extractive methods achieve higher and more stable ROUGE scores than zero-shot generative models. TextRank-Simple and Random-Extract perform best, while all GPT-2 configurations remain substantially lower, and mT5 zero-shot fails to align with references. Decoding parameters exhibit only weak correlations with generative performance, and paired t-tests confirm statistically significant differences (p < 0.05). Overall, extractive approaches remain the most dependable choice without in-domain fine-tuning, while generative models are more suitable with adaptation or hybrid strategies.This study compares extractive and generative approaches for automatic summarization of Indonesian meeting minutes. Our main scientific contribution is an empirical claim that, under strict zero-shot conditions and without domain adaptation, simple extractive baselines are more reliable than off-the-shelf generative models in preserving both decision content and meeting-context cues (actors/roles). We evaluate three extractive baselines (Lead-3, Random-Extract, TextRank-Simple) against an Indonesian GPT-2 model tested under multiple decoding configurations and an mT5 sequence-to-sequence model in a zero-shot setting. Experiments utilize 30 manually curated meeting minutes. The dataset size is intentionally limited because meeting minutes are heterogeneous and require carefully constructed reference summaries to ensure evaluation validity; the study is positioned as a controlled diagnostic comparison rather than a training or adaptation effort. Performance is measured using ROUGE-1/2/L, summary–to–reference length ratios, simple audits of gender and professional role mentions, correlations between decoding parameters and ROUGE, and paired t-tests. Results show that extractive methods achieve higher and more stable ROUGE scores than zero-shot generative models. TextRank-Simple and Random-Extract perform best, while all GPT-2 configurations remain substantially lower, and mT5 zero-shot fails to align with references. Decoding parameters exhibit only weak correlations with generative performance, and paired t-tests confirm statistically significant differences (p < 0.05). Overall, extractive approaches remain the most dependable choice without in-domain fine-tuning, while generative models are more suitable with adaptation or hybrid strategies.
Design IoT for Intelligent Soil Detection in Agriculture and Future Mine-Used Land Reclamation with Mobile Apps Imron, Imron; Satria, Bagus; Ramadhani, Fajar; Karim, Syafei; Dwi Putra Sidik, Rizky
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.v8i1.808

Abstract

Former mining lands in Indonesia, particularly in East Kalimantan, experience severe soil degradation characterized by high acidity, nutrient deficiency, and poor physical structure, which limit their potential for agricultural reuse. This study presents the design and implementation of an integrated Internet of Things (IoT)–based soil monitoring system combined with an artificial intelligence (AI)–driven crop recommendation module to support data-driven land reclamation and precision agriculture. The system consists of an ESP32 microcontroller, NPK soil sensors with RS485 communication, and a cloud-connected mobile application developed using Flutter, Firebase, and ThingsBoard. Soil parameters including pH, moisture, electrical conductivity, temperature, and macronutrients (nitrogen, phosphorus, and potassium) are collected in real time and analyzed using an AI-based reasoning model to generate crop suitability recommendations. System validation was conducted through black-box functional testing covering authentication, data acquisition, geotagged storage, analytics, and recommendation modules. A total of 35 test cases were executed, with 33 cases (94.3%) passing successfully. Performance evaluation shows that dashboard visualization and recommendation generation meet predefined service-level thresholds under normal network conditions. The results indicate that the proposed system is technically feasible for real-time soil monitoring and decision support on post-mining land. However, this study is limited to system-level validation and does not yet include large-scale agronomic field trials or comparative evaluation against conventional soil assessment methods. Future work will focus on improving AI model validation, expanding field deployment, and assessing agronomic impacts over longer cultivation cycles.
Design and Development of a Fisherman's Trap Location Marking Application Using Android-Based Maps Picker and Tracking Kuswanto, Teguh Junian; Hamka, M. Saiful Rahman; Febrianti, Santi; Rokhim, Imam Nur; Nabila, Aisyah Nur
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.v8i1.816

Abstract

The fisheries sector in Indonesia is predominantly composed of small-scale traditional fishermen who still rely on manual methods to determine and track the locations of fishing gear, such as fish traps (bubu). This condition often leads to the loss of fishing gear due to environmental factors such as weather conditions, ocean currents, and limited access to modern navigation technologies. Therefore, there is a need for an information technology–based solution that can assist fishermen in digitally marking and tracking the locations of their fishing gear. This study aims to design and develop an Android-based application capable of marking and tracking the deployment locations of fish traps using a Maps Picker and Global Positioning System (GPS) in real time. The system development adopts the Waterfall model, consisting of requirement analysis, system design, implementation, testing, and evaluation phases. The application is developed using React Native as the primary framework, integrated with the Google Maps API and SQLite as a local database to support offline-first functionality. System testing is conducted using a black-box testing approach. The results indicate that all application features function according to the specified requirements and demonstrate a satisfactory level of tracking accuracy. Therefore, the proposed system is considered effective in assisting fishermen in marking and tracking fishing gear locations efficiently, while also supporting the digital transformation of the small-scale fisheries sector in Indonesia.
Naive Bayes Predictive Model for Failure Detection in CODLAG Propulsion Systems Rokhim, Imam Nur; Kuswanto, Teguh Junian; Dioktyanto, Mudzakkir; Abdullah, Hidayat; Dewi, Yulia Puspa
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.v8i1.831

Abstract

This study implements a Naive Bayes Classifier algorithm to detect failures in gas turbines operating within a CODLAG (Combined Diesel-Electric and Gas) propulsion system. The complexity of hybrid propulsion systems necessitates reliable data-driven monitoring methods to support early anomaly detection and predictive maintenance. An open-access dataset from Kaggle was utilized as the source of gas turbine operational data, with five key parameters (GTn, T48, ṁf, P1, and P2) selected due to their strong correlation with turbine thermodynamic performance. Following data preprocessing and an 80:20 train–test split, the model was trained to classify operating conditions into Normal and Faulty states. The evaluation results demonstrate an accuracy of 86.89%, accompanied by high precision and recall values, indicating the model’s capability to identify anomalies with minimal misclassification. Furthermore, the Receiver Operating Characteristic (ROC) curve yields an Area Under the Curve (AUC) of 0.96, reflecting strong discriminative performance. These findings confirm that the Naive Bayes approach is computationally efficient and suitable for real-time implementation within shipboard Condition-Based Monitoring (CBM) systems, thereby enhancing the reliability and operational efficiency of CODLAG propulsion systems.
Integrating IoT Data and Consumer Behavior Analytics to Enhance Decision-Making in Sustainable Koi Fish Cultivation Sugiarto, Sugiarto; Wahyuni, A; Nugraha, Isna; Rizqina, Azza; Agvenia, Keisya
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.v8i1.832

Abstract

This study presents a data-driven decision-support framework that integrates Internet of Things (IoT)–based water quality monitoring with consumer behavior analytics to support sustainable koi fish cultivation. An IoT monitoring system was implemented using an ESP32 microcontroller equipped with pH, dissolved oxygen (DO), temperature, and turbidity sensors to continuously record water quality conditions over a 30-hour observation period. Time-series sensor data were processed through noise filtering, timestamp synchronization, and descriptive statistical analysis to characterize environmental stability patterns. In parallel, consumer behavior data were collected from 50 respondents using an online questionnaire addressing color preference, purchase considerations, maintenance awareness, and price sensitivity. The integrated analysis combined correlation analysis and K-Means clustering to explore relationships between water quality stability indicators and consumer segmentation. The results indicate that relatively stable pH (6.66–7.20) and DO (6.0–7.1 mg/L) conditions align with the preferences of quality-focused and maintenance-oriented consumer groups, while automated IoT-based monitoring supports operational efficiency relevant to budget-conscious buyers. Overall, the findings demonstrate that integrating environmental sensing data with consumer behavior analytics can enhance operational decision-making, improve market alignment, and support sustainability in koi aquaculture systems.
Waste Classification Using Convolutional Neural Network with ShuffleNetV2 Architecture Fahlevhi, Muhammad Alif; Yoannita, Yoannita
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.v8i1.848

Abstract

Internet of Things (IoT)-based smart waste sorting systems require classification algorithms that are not only accurate but also efficient in resource utilization. However, the majority of previous studies tend to focus on heavy-architecture Deep Learning models (such as VGG or ResNet) that burden edge devices, or utilize lightweight models that are limited to a few class categories. This research contributes to filling this gap by evaluating the effectiveness of the ShuffleNetV2 architecture, a lightweight CNN that optimizes Memory Access Cost (MAC), to classify 9 complex waste categories (Biological, Clothes, Glass, Plastic, Shoes, Battery, Metal, Paper, Cardboard). The research dataset was compiled through the curation and combination of three public Kaggle repositories, which were reprocessed using Roboflow, producing 19,906 augmented images to ensure visual domain variance. Empirical evaluation results show that the model achieved an accuracy of 94% with an average F1-Score of 0.93. The efficiency advantage is evidenced by the compact model size (4.99 MB) and low estimated computational load (0.30 GFLOPs) compared to conventional models. These findings indicate that ShuffleNetV2 offers an optimal performance trade-off, making it a feasible solution for implementation on mobile devices and low-power embedded systems.

Page 22 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