cover
Contact Name
Muhammad Taufiq Nuruzzaman
Contact Email
m.taufiq@uin-suka.ac.id
Phone
+6287708181179
Journal Mail Official
jiska@uin-suka.ac.id
Editorial Address
Teknik Informatika, Fak. Sains dan Teknologi, UIN Sunan Kalijaga Jln. Marsda Adisucipto No 1 55281 Yogyakarta
Location
Kab. sleman,
Daerah istimewa yogyakarta
INDONESIA
JISKa (Jurnal Informatika Sunan Kalijaga)
ISSN : 25275836     EISSN : 25280074     DOI : -
JISKa (Jurnal Informatika Sunan Kalijaga) adalah jurnal yang mencoba untuk mempelajari dan mengembangkan konsep Integrasi dan Interkoneksi Agama dan Informatika yang diterbitkan oleh Departemen Teknik Informasi UIN Sunan Kalijaga Yogyakarta. JISKa menyediakan forum bagi para dosen, peneliti, mahasiswa dan praktisi untuk menerbitkan artikel penelitiannya, mengkaji artikel dari para kontributor, dan teknologi baru yang berkaitan dengan informatika dari berbagai disiplin ilmu
Arjuna Subject : -
Articles 251 Documents
Peningkatan Akurasi Temu Kembali Pengetahuan pada Kasus Sakit Kepala Menggunakan Integrasi Case-Based Reasoning dan Natural Language Processing Agus Mulyanto
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 2 (2026): May 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6189

Abstract

Headache disorders represent a significant global neurological challenge, yet medical diagnosis is often hindered by subjective patient complaints and limited facilities. This study proposes an integrated model combining Natural Language Processing (NLP) and Case-Based Reasoning (CBR) to enhance the accuracy of medical decision-making for headache cases. The model utilizes the Random Forest algorithm for NLP classification and Cosine Similarity within the CBR framework to identify case relevance. The dataset consists of medical records for two types of headaches: Cluster headache (G44.0) and Tension-type headache (G44.2). Experimental results demonstrate that data augmentation significantly improves model performance, increasing accuracy from 62% to 69%. For the G44.0 label, the model achieved a precision of 0.84, while the G44.2 label reached a recall of 0.87. Furthermore, the CBR system strengthens the diagnosis with a similarity level of up to 0.82 and continuous learning capabilities through the Retain stage. This integration effectively provides faster, more targeted diagnostic recommendations for medical professionals.
Tinjauan Sistematis terhadap Arsitektur Convolutional Neural Network untuk Deteksi Kantuk pada Pengemudi Berbasis Visual Michael Sidharta Dharma; Aditiya Hermawan
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.5914

Abstract

This study presents a systematic literature review (SLR) on the application of Convolutional Neural Networks (CNNs) for vision-based driver drowsiness detection. A total of 23 peer-reviewed studies published between 2021 and 2025 were analyzed using the PICOC framework and PRISMA 2020 guidelines. The review aims to identify trends in model architectures, datasets, preprocessing strategies, and performance outcomes. The findings show that transfer learning models, such as MobileNetV2, ResNet, and InceptionV3, along with hybrid architectures like CNN–LSTM, dominate recent research, achieving accuracies predominantly above 90%. However, performance varies significantly depending on dataset characteristics and experimental conditions. Key challenges include sensitivity to low-light environments, occlusion (e.g., glasses), and head pose variations. In addition, most studies rely on controlled datasets, limiting real-world applicability and generalization. Emerging trends emphasize lightweight architectures, temporal modeling, and efficiency-oriented optimization for real-time deployment. This review highlights the need for cross-dataset evaluation, more diverse and representative data, and robust model design to ensure reliable performance in real-world driving scenarios.
Fine-Tuned Artificial Neural Network (ANN) with Explainable AI (XAI) for Classification and Interpretation of the Breast Cancer Wisconsin Ikra Muhibban Ananta; Hanif Amarudin; Restu Adi Akbar; M. Fajratul Ikhsan; Maria Ulfah Siregar
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.5973

Abstract

Breast cancer remains one of the leading causes of cancer-related mortality among women, making early detection and accurate diagnosis essential. Although machine learning and deep learning approaches have achieved high classification performance, their interpretability and evaluation robustness remain limited. This study proposes a Fine-Tuned Artificial Neural Network (ANN) integrated with Explainable Artificial Intelligence (XAI) for breast cancer classification using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. Model evaluation was conducted using Stratified 5-Fold Cross-Validation, while optimization was performed through learning rate adjustment and early stopping. Model interpretability was analyzed using SHapley Additive exPlanations (SHAP), including summary, decision, force, and waterfall plots. Experimental results show that the baseline ANN achieved an average accuracy of 97.54%, precision of 97.82%, recall of 98.32%, and F1-score of 98.04%. The Fine-Tuned ANN improved performance, achieving 97.89% accuracy, 98.08% precision, 98.60% recall, and 98.32% F1-score, with better stability across folds. SHAP analysis identified worst area, worst radius, worst texture, and mean concave points as the most influential features. Furthermore, the ablation study indicated that the Sigmoid activation function, Adam optimizer, and learning rate of 0.01 produced the optimal configuration.
An Ethical Information System Success Framework for Government Artificial Intelligence Dwi Yuniarto; A'ang Subiyakto; Aedah Abd. Rahman
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6062

Abstract

The rapid adoption of Artificial Intelligence (AI) in government has transformed public service delivery and decision-making while raising ethical concerns related to fairness, transparency, accountability, and privacy. This study develops and validates an Ethical Information System Success Framework for Artificial Intelligence in Government by integrating Ethical AI governance principles with the Information System Success Model. A sequential mixed-method approach with triangulation-based instrument testing was employed. The model was developed using an Input–Process–Output logic and operationalized into 11 latent variables with 55 measurement items. Quantitative data were collected from 86 government employees using AI-based systems during October–November and analyzed using PLS-SEM, complemented by qualitative expert validation. The findings reveal that ethical transparency, accountability, and privacy and security significantly influence user trust, whereas ethical fairness does not demonstrate a statistically significant effect. User trust and satisfaction subsequently drive actual AI use, which strongly contributes to net public benefits. The empirical study was conducted among Indonesian government employees using AI-supported public service systems.
Deteksi Anomali Jaringan Transaksi Token ERC20 di Blockchain Ethereum Menggunakan Graph Neural Network Nazhif Alaudin Fahmi; Andrey Ferriyan
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6065

Abstract

The rapid development of Decentralized Finance (DeFi) on the Ethereum blockchain has increased ERC20 token transaction volumes but also triggered a surge in anomalous activities such as Ice Phishing, NFT Scams, and Reward Scams. Conventional detection methods on tabular data often fail to capture complex relational patterns between accounts. This study proposes a Graph Neural Network (GNN) approach using the GraphSAGE architecture to detect anomalies in ERC20 transaction networks. Transaction datasets were acquired from Etherscan, processed into graphs with additional structural features (PageRank, degree), and imbalance was handled using class weighting. Based on testing using 33,848 transactions, the model yielded an average accuracy of 95.92% and a macro F1-score of 82.62%. These results prove that GraphSAGE is effective in learning feature representations and graph structures to detect various ERC20 fraud modes simultaneously.
Penerapan Convolutional Block Attention Module untuk Meningkatkan Klasifikasi Glaukoma Menggunakan ResNeXt Chalifa Chazar; Arib Arkaan Fadhlullah; Milda Gutiana Husada; Uung Ungkawa
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6084

Abstract

Glaucoma is a chronic ocular disease characterized by progressive optic nerve damage and is one of the leading causes of irreversible blindness. Manual analysis of retinal fundus images is time-consuming, particularly under high clinical workloads. To improve the efficiency of clinical diagnostic workflows, this study aims to develop a glaucoma classification system using a hybrid deep learning model that integrates the ResNeXt architecture with a Convolutional Block Attention Module (CBAM). In the proposed approach, ResNeXt is employed to capture complex feature representations, while CBAM enhances feature extraction by directing the model’s attention to more informative spatial and channel-wise features. Performance evaluation demonstrates that the proposed model outperforms the baseline architecture. The standard ResNeXt model achieves an accuracy of 91.61%, an F1-score of 88.98%, and an AUC of 93.1%. In contrast, the ResNeXt model integrated with CBAM shows improved performance, attaining an accuracy of 92.70%, an F1-score of 90.60%, and an AUC of 95,6%. These results indicate that the integration of CBAM effectively enhances the sensitivity and robustness of the glaucoma classification model compared to the standard architecture.
Enhanced Ant Colony Optimization for Cloud Scheduling with Local Search and Elitist I Made Alit Darma Putra
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6105

Abstract

Cloud computing allows the execution of various types of jobs with diverse resource requirements, so an efficient scheduling mechanism is needed to minimize processing time and improve resource utilization. The Ant Colony Optimization (ACO) algorithm is one of the metaheuristic methods widely used for job scheduling optimization. However, the ACO algorithm still has limitations in slow convergence and a tendency to get stuck in local solutions. This study proposes Enhanced Ant Colony Optimization with the Swap-Based Local Search and Elitist Strategy approaches to improve the quality of scheduling decisions. Evaluations were conducted in a heterogeneous emulated cloud environment with a focus on scheduling performance, convergence behaviour, and resource utilization metrics. The experimental results show that the proposed method outperforms FCFS, ACO, and ACO with Local Search for both normal and heavy scenarios. The study also demonstrates that the integration of local search and elitist strategy effectively improves the convergence process toward the global optimal solution while promoting a fairer utilization of resources. This paper may serve as a basis for further development on larger job scales and the addition of other optimization techniques.
Analisis Komparatif Akurasi dan Stabilitas Model Peramalan Harga Beras Harian Nasional Restu Puji Hidayat
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6149

Abstract

Daily forecasting of national rice prices is important for supporting data-driven food price monitoring systems. This study evaluates statistical, machine learning, and deep learning approaches for forecasting Indonesia’s national rice prices over the 2017–2025 period. Missing values in the daily data were handled using the Last Observation Carried Forward (LOCF) method, limited to gaps of up to seven days, which successfully filled all missing observations. The completed time series was then transformed into a supervised learning format using a seven-day sliding window selected through temporal analysis and validation-based sensitivity testing. The dataset was divided chronologically, with the final 90 days reserved as the test set. The evaluated models included Seasonal Naive, XGBoost, Support Vector Regression optimized using Particle Swarm Optimization (SVR-PSO), and Temporal Convolutional Network (TCN). Model performance was evaluated using MAE, RMSE, MAPE, and MSE, while SVR-PSO and TCN were each run 10 times to assess performance stability. SVR-PSO achieved the best performance, with an MAE of 16.830 ± 2.722, an RMSE of 22.189 ± 2.216, a MAPE of 0.1068% ± 0.0173, and an MSE of 497.270 ± 107.992. These findings indicate that SVR-PSO produces relatively accurate and stable short-term forecasts and has the potential to support the development of data-driven national food price monitoring and forecasting systems.
Improving Multi-Class Classification Performance with Multi-Scale Feature Extraction CNNs on a Limited Indonesian Spices Dataset Ahmad Zein Al Wafi; Alena Ghinna Kirana
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6185

Abstract

This study evaluates whether multi-scale feature extraction provides an advantage over single-scale CNN feature extraction under limited-data conditions. Using the Indonesian Spices Dataset as a controlled benchmarking testbed, which contains 31 distinct categories with only 210 images per class, we propose a multi-scale Convolutional Neural Network (CNN) architecture that captures features across varying spatial scales through three parallel branches with 3x3, 5x5, and 7x7 kernel sizes. The proposed model was evaluated against single-scale baselines using training and validation accuracy and loss, macro-averaged precision, recall, F1-score, and a confusion matrix. Experimental results show that the multi-scale approach provides the strongest learning capacity among the evaluated architectures, achieving the highest training accuracy (91.44%) and the lowest training loss (0.2713), while reaching a peak validation accuracy of 64.98% (62.90% final-epoch accuracy, 64.22% macro precision, and 62.59% macro F1-score) with a competitive training time of 3839.7s. The validation-accuracy advantage over the strongest single-scale baseline is modest; the clearer advantage is observed in training capacity and the broader validation metric profile. The late-epoch divergence between training and validation loss indicates that the additional representational capacity remains constrained by the limited number of training samples, highlighting the need for stronger regularization to improve generalization in scarce multi-class data.
Integrasi Citra Satelit dan Inverter Berbasis Transfer Learning untuk Prediksi PLTS Tropis Petricia Oktavia; Tole Sutikno; Murinto Murinto
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 11 No. 3 (2026): September 2026
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.6297

Abstract

Integrating photovoltaic (PV) power plants into smart grids requires reliable short-term forecasting, particularly under highly variable tropical atmospheric conditions. Conventional univariate temporal models, such as LSTM, struggle to anticipate abrupt cloud cover changes as they rely solely on historical electrical data. This study proposes a hybrid multimodal CNN-LSTM framework integrating 16-channel Himawari-9 spatio-temporal satellite imagery and real-time inverter electrical parameters via cross-climate transfer learning and Solar Polar Image Normalization (SPIN). Multimodal alignment was achieved using linear temporal and bilinear spatial interpolations across 9,622 synchronized 5-minute samples from Universitas Pamulang, while a dedicated multimodal adapter enabled partial weight transfer from the Stanford SKIPP’D dataset. Experimental results demonstrate superior performance, achieving an MAE of 3,73 kW, RMSE of 5,61 kW, a Forecast Skill of +5,98%, and mitigating extreme Q95 error to 12,37 kW compared to persistence and univariate LSTM baselines. With a CPU inference latency of 15,58 ms per sample, this architecture provides high accuracy and computational efficiency for real-time tropical grid operations.