cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Dede Kurniadi
Contact Email
dede.kurniadi@itg.ac.id
Phone
+6287880007464
Journal Mail Official
jistics@aptika.org
Editorial Address
Green Garden Residence C-87, Kabupaten Garut, Provinsi Jawa Barat, Indonesia, 44151
Location
Kab. garut,
Jawa barat
INDONESIA
Journal of Intelligent Systems Technology and Informatics
Published by Aliansi Peneliti Informatika
ISSN : -     EISSN : 3109757X     DOI : https://doi.org/10.64878/jistics
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Engineering
The Journal of Intelligent Systems Technology and Informatics (JISTICS) is an international peer-reviewed open-access journal that publishes high-quality research in the fields of Artificial Intelligence, Intelligent Systems, Information Technology, Computer Science, and Informatics. JISTICS aims to foster global scientific exchange by providing a platform for researchers, practitioners, and academics to disseminate original findings, critical reviews, and innovative applications. The journal is published three times a year (March, July, November) and may also publish special issues on emerging topics.
Arjuna Subject : Ilmu Komputer (semua kategori) - Aplikasi Ilmu Komputer
Articles 31 Documents
 1   2   3   4 
Analysis of Earthquake Notification Complaint Topics in Info BMKG Reviews Using BERTopic Hanipah Diniyaturobiah; Rifky Khoerul Muzaky
Journal of Intelligent Systems Technology and Informatics Vol 2 No 1 (2026): JISTICS, March 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i1.123

Abstract

Reliable earthquake notification services in public information applications play a critical role in supporting public awareness and preparedness in seismically active regions. This study examines user complaints about earthquake notification features in the Info BMKG mobile application by analyzing publicly available Google Play Store user reviews. A total of 1,500 reviews were collected and examined, with complaint reviews operationally defined as those with star ratings of 3 or lower. Prior to analysis, the dataset underwent text preprocessing and a balancing procedure to ensure adequate representation of complaint-related content. Topic modeling was conducted using BERTopic, a transformer-based approach that enables context-aware clustering of short, informal text, followed by descriptive temporal analysis to examine variations in complaint occurrence over time. The analytical workflow included text normalization, embedding generation, topic extraction, and temporal mapping of complaint patterns. The results reveal several recurring complaint themes, including delayed or missing notifications, clarity of information, application performance issues, and user responses to system updates. Temporal variations indicate periods of increased complaint activity that align with heightened application usage, reflecting shifts in user engagement rather than direct evidence of system failure. Topic validity was assessed through qualitative inspection of representative reviews to ensure semantic consistency and interpretability. Overall, this study provides a structured, descriptive overview of user concerns regarding earthquake notification services and demonstrates the applicability of topic-level and temporal analysis as an evaluative approach for mobile disaster information applications, without making causal performance claims.
Transaction Segmentation of Supermarket Sales Data for Retail Decision Support Using K-Means Clustering Fakhrun Mahda Khoiriyyah; Hery Suhendar; Yusep Maulana
Journal of Intelligent Systems Technology and Informatics Vol 2 No 1 (2026): JISTICS, March 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i1.145

Abstract

The increasing availability of transactional data in the retail sector provides opportunities to support data-driven managerial decision-making. This study aims to segment supermarket sales transactions using the K-Means clustering method to identify meaningful transaction patterns that support retail decision-making. A publicly available supermarket transaction dataset was analyzed using selected numerical attributes representing purchase quantity, transaction value, and customer rating. To ensure reliable and interpretable clustering results, data standardization was applied, and the optimal number of clusters was determined using a combined validation strategy comprising the Elbow Method and the Silhouette Score. The results indicate that three distinct transaction segments were identified, characterized by similar purchase quantities but differing transaction values and customer satisfaction levels. Principal Component Analysis visualization confirms that the resulting clusters are well separated and interpretable. The findings demonstrate that integrating systematic cluster validation with interpretable cluster analysis provides practical insights for retail managers in designing targeted marketing strategies, improving customer satisfaction, and supporting inventory and promotional decision-making.
Rethinking Efficiency: A Comparative Study of Lightweight CNN Architectures for Image Classification Mochamad Rizal Fauzan; Naufal Nadhif Rabbani Iskandar; Rafi Zahran Fauzi
Journal of Intelligent Systems Technology and Informatics Vol 2 No 1 (2026): JISTICS, March 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i1.167

Abstract

Lightweight convolutional neural networks (CNNs) are increasingly required for image classification in resource-constrained environments; however, their comparative behavior under unified training conditions remains insufficiently explored, particularly when accuracy, parameter efficiency, inference latency, and augmentation sensitivity are evaluated simultaneously. This study presents a systematic benchmark of five lightweight CNN architectures, namely MobileNetV2, EfficientNet-B0, ShuffleNetV2, SqueezeNet, and ResNet18, on the CIFAR-100 dataset using a consistent experimental pipeline. All models were trained for 40 epochs with an input resolution of 128 × 128, AdamW optimization, cosine annealing, mixed-precision training, and identical preprocessing settings. Two augmentation strategies, namely basic and advanced augmentation, were evaluated to examine their influence on model generalization. The results show that EfficientNet-B0 achieved the best classification performance with 82.75% Top-1 accuracy and 96.46% Top-5 accuracy, while SqueezeNet achieved the fastest inference latency of 1.52 ms and the smallest parameter size, indicating its suitability for highly constrained deployment scenarios. Across all evaluated models, the average Top-1 and Top-5 accuracies reached 76.6% and 94.16%, respectively. In addition, the effect of advanced augmentation was found to be architecture-dependent rather than uniformly beneficial. On average, it resulted in a Top-1 accuracy change of −0.66 percentage points, with only ResNet18 showing a modest improvement. The main contribution of this study is to provide a unified, practically oriented benchmark that highlights how architectural design, rather than parameter count alone, determines the balance between accuracy and computational efficiency. These findings provide clearer guidance for selecting lightweight CNN models for real-world image classification tasks under varying deployment constraints.
Empirical Benchmarking of Hybrid Retrieval in Educational Conversational AI: Accuracy‑Latency Trade‑offs and Robustness Abdul Saboor Hamedi; Iqbal Hussain Alamyar; A.A. Waskita
Journal of Intelligent Systems Technology and Informatics Vol 2 No 2 (2026): JISTICS, July 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i2.245

Abstract

This study investigates the comparative performance of lexical, semantic, and hybrid retrieval strategies in educational conversational AI, with a focus on accuracy-latency trade‑offs and robustness across diverse query types. A controlled experimental framework was implemented using PostgreSQL’s ts_rank for lexical retrieval, pgvector embeddings for semantic retrieval, and two fusion strategies: Linear Weighted Fusion and Reciprocal Rank Fusion. The evaluation corpus consisted of approximately 50,000 text chunks extracted from 2025 arXiv AI/ML papers, and a benchmark of 100 queries spanning conceptual, factual, procedural, comparative, and miscellaneous categories was executed. Effectiveness was measured using NDCG@10, Precision@5, and MRR, while efficiency was quantified via end‑to‑end latency. Relevance judgments were generated through an AI‑as‑a‑Judge pipeline to ensure scalability and reproducibility. Results showed that semantic and hybrid methods achieved a high accuracy mean NDCG@10 ≈ 0.91 but incurred latency costs between 227-505 ms. Lexical retrieval was fastest, 88 ms, but substantially less accurate, 0.346. Hybrid‑Linear fusion emerged as the most robust strategy, winning 66% of queries in the Winner‑Take‑All analysis, while semantic search excelled in conceptual queries and lexical search in acronym‑based factual lookups. Reciprocal Rank Fusion achieved comparable mean accuracy but failed to dominate in any category. The findings highlight a clear quality–speed dichotomy and establish Hybrid‑Linear fusion as the most dependable retrieval method for educational chatbots. For latency‑sensitive applications, semantic search offers the best balance of responsiveness and accuracy. The study provides actionable design guidelines and identifies future directions, including corpus generalization, human evaluation calibration, intelligent query routing, and latency optimization.
Skin Disease Classification on the Body Area Using a Combination of Convolutional Neural Network and Vision Transformer Yogi Sugiman; Muhammad Daffa Adzdzikra Daniswara
Journal of Intelligent Systems Technology and Informatics Vol 2 No 2 (2026): JISTICS, July 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i2.186

Abstract

Skin diseases affecting the body area represent a significant dermatological challenge due to the high visual similarity between conditions, which complicates accurate diagnosis. Prior studies have predominantly relied on pure CNN architectures, which are inherently limited in capturing long-range contextual relationships between distant lesion regions, and no existing work has integrated CNN with Vision Transformer specifically for body area skin disease classification. This study addresses this gap by applying the SEMMA (Sample, Explore, Modify, Model, Assess) methodology to develop a novel hybrid classification model integrating Convolutional Neural Network (CNN) and Vision Transformer (ViT) for classifying four types of skin diseases on the body area, namely Acne and Rosacea, Eczema, Psoriasis, and Tinea Ringworm and Fungal Infections, using the DermNet dataset. Two CNN backbones were evaluated: DenseNet201 and EfficientNetB4, each combined with four Pre-LayerNorm Transformer Blocks that feature learnable positional encoding to capture long-range spatial dependencies among feature tokens. A two-phase training strategy was implemented, consisting of feature extraction followed by fine-tuning, with balanced class weight applied to address class imbalance detected during exploratory data analysis. The original training data was split into training and validation sets at a 90:10 ratio, while the test data was sourced from the built-in test folder of the DermNet repository, yielding 4,301 training images, 479 validation images, and 1,298 test images. Evaluation on fully isolated test data showed that DenseNet201+ViT achieved 78% accuracy and EfficientNetB4+ViT achieved 77% accuracy. The application of the probability-averaging ensemble strategy further improved performance to 82% accuracy and a Macro F1-Score of 82%, surpassing prior CNN-based studies in the same domain. Grad-CAM visualization confirmed that the model focused attention on clinically meaningful lesion areas, indicating that its predictions are grounded in relevant morphological features rather than image artifacts. However, this study is limited by the absence of clinical validation on real patient data, and the current accuracy remains below the threshold required for standalone diagnostic use. The findings demonstrate that hybrid CNN+ViT architectures, combined with ensemble strategies, offer a promising and interpretable approach for automated skin disease classification, warranting further clinical validation before deployment as a diagnostic support tool.
Comparative Analysis of Vision Transformer (DeiT) and Transfer Learning-Based CNN (ResNet50) Performance in Identifying AI-Generated 2D Animation Illustrations Hasan; Dimas Hendra Yudha; Nazwa Mutia Salma
Journal of Intelligent Systems Technology and Informatics Vol 2 No 2 (2026): JISTICS, July 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i2.213

Abstract

The development of generative artificial intelligence has introduced new challenges in the digital creative industry, particularly in distinguishing human-created 2D anime illustrations from AI-generated images due to their increasingly similar visual characteristics. This study aims to compare the performance of a Vision Transformer model, the Data-efficient Image Transformer (DeiT), and a transfer-learning-based Convolutional Neural Network model, ResNet50, for detecting AI-generated 2D anime illustrations. The study employed the SEMMA methodology, consisting of the Sample, Explore, Modify, Model, and Assess stages. An initial dataset of 2,000 images was collected from Safebooru and Civitai, of which 1,736 images were validated and used in the experiments. The images underwent preprocessing, including resizing, normalization, data augmentation, and splitting the dataset into training, validation, and test sets. Model performance was evaluated using accuracy, precision, recall, F1-score, confusion matrix, and ROC-AUC metrics. The experimental results showed that ResNet50 achieved 94% accuracy with an AUC of 0.993, while DeiT achieved a slightly higher accuracy of 95% with an AUC of 0.990. Although both models demonstrated excellent discriminative power, DeiT achieved slightly better overall classification performance under the same experimental setting. These findings indicate that Vision Transformer-based models have strong potential for detecting AI-generated anime illustrations and can contribute to the development of more reliable digital artwork authenticity detection systems.
Optimizing Daily Household Energy Consumption Prediction Using Ensemble Machine Learning, Feature Selection, and Explainable Artificial Intelligence Bubu Bukhori Muslim; Jujun Munawar
Journal of Intelligent Systems Technology and Informatics Vol 2 No 2 (2026): JISTICS, July 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i2.224

Abstract

Household energy consumption has become an important issue due to the increasing global energy demand and the need for efficient energy management. Accurate prediction of household energy consumption can support energy planning, reduce energy waste, and improve decision-making in residential energy management. However, developing prediction models that achieve both high predictive performance and interpretability remains a challenging task. Therefore, this study aims to optimize household energy consumption prediction by integrating Ensemble Machine Learning, Feature Selection, and Explainable Artificial Intelligence (XAI). The proposed framework follows the Cross-Industry Standard Process for Data Mining (CRISP-DM), comprising business understanding, data understanding, data preparation, modeling, evaluation, and explainability analysis. The dataset used in this study contains 90,000 household energy consumption records. Feature selection was performed using XGBoost feature importance, while Random Forest, XGBoost, Gradient Boosting, Voting Regressor, and Stacking Regressor were employed as predictive models. Model performance was evaluated using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), coefficient of determination (R²), and 10-fold cross-validation. The experimental results indicate that XGBoost achieved the best performance with an MAE of 0.5108, RMSE of 0.6725, MAPE of 0.0607, and R² of 0.9852. Furthermore, SHAP analysis revealed that Peak_Hours_Usage_kWh and Household_Size were the most influential features affecting household energy consumption. In conclusion, the integration of Ensemble Machine Learning, Feature Selection, and XAI effectively yields an accurate, robust, and interpretable model for predicting household energy consumption.
FedProx Optimization Based on Compression and Quantization for Efficient Heart Disease Prediction on Edge Devices Anyelir Kuntum Sari; Tabina Athifa Rahmaniya
Journal of Intelligent Systems Technology and Informatics Vol 2 No 2 (2026): JISTICS, July 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i2.236

Abstract

Cardiovascular disease remains a leading cause of mortality worldwide, posing challenges for early detection within privacy-sensitive medical environments. Federated Learning (FL) offers a promising solution by enabling distributed model training without transferring raw patient data to a central server. However, FL deployment on edge devices is constrained by limited resources, system heterogeneity, and non-IID data, which degrade model stability and convergence. This study integrates sparsification-based compression and 16-bit quantization into FedProx and evaluates their combined and individual effects on heart disease prediction under compounded non-IID and straggler conditions. Using the Heart Disease Train-Test dataset from the UCI repository (1,025 samples, 14 attributes) and a KDD methodology, experiments simulated a non-IID FL environment with 10 clients and 30% stragglers over 50 rounds, averaged over five seeds. Results show that standard FedProx achieves the highest mean performance (86.44% accuracy, 87.34% F1-Score), closely followed by FedAvg (86.15%, 87.18%), while the proposed method attains a lower mean accuracy (84.88%) and F1-Score (85.87%), not statistically significant (p = 0.061, 0.098). ROC-AUC, however, is significantly lower for the proposed method (93.52% vs. 94.42%, p < 0.01), indicating reduced discriminative ability. An ablation study shows compression, not quantization, primarily drives the efficiency gains, reducing transmitted parameter size by 45.09% and accelerating convergence to 20.2 rounds compared with the baseline methods. These findings indicate that integrating compression and quantization into FedProx achieves substantial communication efficiency and faster convergence with only minor performance trade-offs, offering a viable option for resource-constrained edge deployments.
Fake News Detection in Indonesian Language Using IndoBERT with LIME-Based Keyword Interpretation Rifki Ramdani; Muhammad Nadhief Rahmat Firdaus
Journal of Intelligent Systems Technology and Informatics Vol 2 No 2 (2026): JISTICS, July 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i2.198

Abstract

Fake news dissemination in the digital era has become a serious issue, particularly in political and public information domains. This study proposes an Indonesian fake news detection system that uses the IndoBERT transformer model, combined with LIME (Local Interpretable Model-Agnostic Explanations), for keyword-based interpretation. The primary objective of this study is not only to achieve high classification performance but also to enhance model transparency by identifying the most influential words contributing to prediction results. This study follows the SEMMA (Sample, Explore, Modify, Model, Assess) methodology, starting with dataset collection, exploratory data analysis, text preprocessing, model fine-tuning, and evaluation, and concluding with interpretability analysis using LIME. The dataset consists of 31,310 Indonesian political news articles categorized into hoax and factual classes. IndoBERT is fine-tuned using the Hugging Face framework with optimized hyperparameters and class weighting to address class imbalance. Experimental results show that the proposed model achieves an accuracy of 99.78%, precision of 99.81%, recall of 99.52%, and F1-score of 99.66%, demonstrating strong performance in distinguishing hoax and factual news. Furthermore, LIME-based analysis provides interpretable insights by highlighting keywords that influence model predictions, thereby improving transparency and user trust. Words associated with conspiracy and unverified claims contribute strongly to the hoax class, while terms related to official institutions and statistical information support factual classification. The results indicate that integrating IndoBERT with LIME not only improves classification performance but also enhances explainability in Indonesian fake news detection systems.
Adverse Price Excursion Risk Prediction for Margin-Call Early Warning in Forex Trading Using Attention-BiLSTM Across Currency Pairs Gea Natasya; Nandang Taufik WM
Journal of Intelligent Systems Technology and Informatics Vol 2 No 2 (2026): JISTICS, July 2026
Publisher : Aliansi Peneliti Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64878/jistics.v2i2.259

Abstract

Leveraged foreign-exchange trading is exposed to rapid adverse price movements that can contribute to margin-call events, yet most prior studies emphasize price or direction forecasting rather than early risk classification. This study uses the Sample, Explore, Modify, Model, and Assess (SEMMA) framework to organize a time-aware experiment on hourly EUR/USD and GBP/USD data from 2010 to 2026. After cleaning, the datasets contained 99,987 and 99,985 observations, respectively. A direction-agnostic maximum adverse excursion proxy labeled whether either a hypothetical long or short position would experience at least 50 pips of adverse movement within five hours. Sixteen technical features were converted into 60-step sequences and evaluated using identical five-fold chronological walk-forward splits. XGBoost, BiLSTM, Attention-BiLSTM, TransformerEncoder, and PatchTST-Lite were compared using training-only scaling and validation-only threshold selection. XGBoost achieved the strongest mean F1-score and ROC-AUC on EUR/USD (0.3710 and 0.7927) and GBP/USD (0.4855 and 0.7847). Attention-BiLSTM remained competitive, with F1-scores of 0.3583 and 0.4825 and the highest mean recall on GBP/USD (0.6680). In a no-retraining EUR/USD-to-GBP/USD transfer test, it obtained an F1-score of 0.5152 and ROC-AUC of 0.7923. Five-fold Wilcoxon tests lacked sufficient resolution to establish superiority. The results support Attention-BiLSTM as a temporally attributable early-warning component, while XGBoost offers the best efficiency-performance trade-off.

Page 3 of 4 | Total Record : 31

 1   2   3   4 

Filter by Year

2025 2026


Reset
Filter By Issues
All Issue Vol 2 No 2 (2026): JISTICS, July 2026 Vol 2 No 1 (2026): JISTICS, March 2026 Vol 1 No 3 (2025): JISTICS, November 2025 Vol 1 No 2 (2025): JISTICS, July 2025 Vol 1 No 1 (2025): JISTICS, March 2025