cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Much Aziz Muslim
Contact Email
a212muslim@mail.unnes.ac.id
Phone
+628164243462
Journal Mail Official
adminjoiser@shmpublisher.com
Editorial Address
SHM Publisher No. 64, Karanglo st, Pedurungan Distr, Semarang, Central Java, Indonesia 50191.
Location
Kota semarang,
Jawa tengah
INDONESIA
Journal of Information System Exploration and Research
Published by shm publisher
ISSN : 29641160     EISSN : 29636361     DOI : https://doi.org/10.52465/joiser.xxxx
Core Subject :
Journal of Information System Exploration and Research is a journal that publishes and disseminates scientific research papers on information systems to a wide audience particularly within the information system society
Arjuna Subject : -
Articles 7 Documents
 1 
Search results for , issue "vol. 4 no. 4: october 2026" : 7 Documents clear
E-Commerce Dynamic Pricing Optimization Using Double Deep Q-Network and Reward Shaping NAZAL Mahendra; Majid Rahardi
Journal of Information System Exploration and Research Vol. 4 No. 4: October 2026
Publisher : shmpublisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joiser.v4i4.90

Abstract

Dynamic pricing is crucial for e-commerce platforms to adapt to rapidly fluctuating market demands. However, conventional reinforcement learning methods, such as Tabular Q-Learning, suffer from the curse of dimensionality and discretization errors. Furthermore, the Deep Q-Networks (DQN) standard often encounters overestimation bias, leading to aggressive price drops and severe financial losses known as the Death Spiral. This study aims to address these issues by proposing a continuous Markov Decision Process (MDP) environment using a Double Deep Q-Network (DDQN) architecture. The model is trained using historical time-series data from the Brazilian E-Commerce Public Dataset by Olist. The proposed innovations include internal State Normalization to eliminate Gradient Domination, and adaptive Reward Shaping paired with absolute economic Guardrails to strictly prevent the AI agent from setting prices below the base cost. Comparative evaluation results demonstrate that the DDQN agent achieved a highly stable convergence, generating a maximum cumulative profit of IDR 11,146,669. This proposed model significantly outperformed the static pricing baseline (IDR 3,975,383) and the conventional Q-Learning algorithm (IDR 9,966,507). These findings prove the robustness of the proposed DDQN architecture in safely, precisely, and automatically exploiting continuous market inelasticity without relying on manual interventions.
Comparison of CNN and HOG-Based Machine Learning Methods for AI-Generated Image Classification Using the CIFAKE Dataset Dika Kurnia Putra; Chandra Ari Pradana; Muhammad Hilal Gilardin; Jordy Lasmana Putra
Journal of Information System Exploration and Research Vol. 4 No. 4: October 2026
Publisher : shmpublisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joiser.v4i4.92

Abstract

The rapid development of Generative Artificial Intelligence (Generative AI) has enabled the creation of highly realistic images that are increasingly difficult to distinguish from real images. This raises concerns about misinformation, identity forgery, and digital media manipulation, highlighting the need for reliable image classification methods. This study compares Convolutional Neural Network (CNN) features with Histogram of Oriented Gradients (HOG)-based machine learning methods, including K-Nearest Neighbor (KNN), Decision Tree (DT), and Random Forest (RF), for classifying AI-generated and real images. The CIFAKE dataset, consisting of real and AI-generated images, was used in the experiments. CNN performed automatic feature extraction, while KNN, DT, and RF used HOG feature vectors. Model performance was evaluated using accuracy and confusion matrix analysis. The results show that CNN achieved the highest accuracy of 87.30%, followed by Random Forest (75.00%), KNN (72.27%), and Decision Tree (63.98%). These findings indicate that CNN is more effective in capturing visual patterns in AI-generated images, while HOG-based methods offer lower computational complexity. Therefore, CNN provides better classification performance for AI-generated image detection.
Balanced Cluster Sampling for Data-Efficient DNN Air-Quality Classification on a Synthetic Urban Sensor Archive Mohammed Hameed
Journal of Information System Exploration and Research Vol. 4 No. 4: October 2026
Publisher : shmpublisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joiser.v4i4.98

Abstract

Large-scale urban sensing platforms generate millions of environmental observations, making full-data DNN training computationally costly. This paper presents a controlled empirical study of balanced cluster sampling (BCS), which partitions preprocessed feature space and draws equal quotas under a fixed instance budget. BCS is compared with uniform random sampling, label-stratified sampling, and two established coreset selectors using a two-million-row synthetic archive modeled on the Array of Things Chicago deployment. The pipeline applies log1p transformation, train-only z-score normalization, cyclical and one-hot calendar encoding, and a fixed ten-layer DNN. On a stratified test set of 300,000 rows, the full-data baseline achieves 98.01 ± 0.27% accuracy. BCS with K = 8 achieves 96.72 ± 0.55% accuracy and 96.29 ± 0.60% macro-F1 using 14.3% of training data, compared with 96.52 ± 0.64% for uniform random sampling; the 0.19-point difference is not statistically significant (p = 0.532). Label-stratified sampling performs at least as well (97.05 ± 0.41%), while Moderate Coreset and EL2N achieve 89.03 ± 1.35% and 55.54 ± 11.91%, respectively. Although pruning reduces DNN training time by 71.0%, end-to-end time decreases by only 43.3%. These results indicate that, for this workload, uniform or label-stratified sampling is sufficient, while additional selection machinery provides no practical benefit.  
A White-Label Microservices Platform for Multi-Institution Smart Village: A Case Study of PT Citiasia, Inc. Muhammad Raihan Alfaiz; Safira Faizah; Halimatuz Zuhriyah
Journal of Information System Exploration and Research Vol. 4 No. 4: October 2026
Publisher : shmpublisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joiser.v5i1.100

Abstract

Village-service digitalization in Indonesia is often constrained by fragmented applications and repetitive redevelopment for each institution. This study investigates a White-Label Smart Village platform at PT Citiasia, Inc. with two objectives: enabling dynamic institutional branding without source-code modification and evaluating a refactored Microservices prototype for multi-institution deployment. A qualitative case-study approach drew on six expert informants, regulatory documents, historical project information, and technical artifacts. The industrial MVP was developed as a modular monolith from August to December 2025 and refactored in February-March 2026 into independently deployable Laravel services running in separate Docker containers. NGINX serves as the API Gateway, while service-owned databases and a Database-per-Tenant strategy provide physical separation between institutions. Functional validation covered 29 Black Box scenarios. Sequential local-staging measurements produced baseline API response times of 175-854 ms. Runtime White-Label configuration changed institutional branding without rebuilding the application, while a single staging trial reduced onboarding to approximately 15 minutes compared with a historical 4-6 working-day estimate. Because production-scale concurrent load and formal penetration testing were outside the study scope, scalability and security claims are limited to architectural and preliminary evidence.
Non-Invasive Anemia Screening Using Fingernail Images with EfficientNetV2-B0 Maulany Citra Pandini; Nurul Huda
Journal of Information System Exploration and Research Vol. 4 No. 4: October 2026
Publisher : shmpublisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joiser.v4i4.111

Abstract

Anemia is a prevalent health condition requiring timely detection for appropriate clinical management. Conventional diagnosis relies on blood-based examinations, which are invasive and may be less practical for rapid and accessible screening. These limitations highlight the need for non-invasive and automated approaches to support preliminary anemia screening. This study develops and evaluates a non-invasive automated anemia classification framework using fingernail images and EfficientNetV2-B0. A dataset of 4,260 fingernail images from anemia and non-anemia classes was used. The images were converted into RGB format and resized to a resolution of 224 × 224 pixels, normalized using ImageNet statistics, and augmented to improve robustness. EfficientNetV2-B0 was pre-trained on ImageNet and subsequently fine-tuned to perform binary classification. The model achieved 97.89% accuracy, 0.98 precision, 0.97 recall, 0.98 F1-score, and 0.99 ROC-AUC on the independent test set. These findings suggest that the proposed framework could support rapid and non-invasive automated anemia screening.
Pneumonia Classification from Chest X-Rays Using CLAHE and Xception-Based Transfer Learning Adi fajar saputra; Nurul Huda
Journal of Information System Exploration and Research Vol. 4 No. 4: October 2026
Publisher : shmpublisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joiser.v4i4.112

Abstract

Pneumonia classification from chest X-ray (CXR) images remains challenging because radiographic appearance varies in contrast, intensity, anatomical structure, and disease manifestation, while class imbalance may affect automated classification. This study proposes a transfer learning framework based on the Xception architecture for binary pneumonia classification from CXR images. The proposed framework integrates systematic image preprocessing, including image resizing, Contrast Limited Adaptive Histogram Equalization (CLAHE), Xception-specific normalization, training-only data augmentation, and class weighting. The dataset consisted of 5,856 CXR images and was stratified into 70% training, 20% validation, and 10% testing subsets. Xception was trained using a two-phase strategy, comprising 20 epochs with the backbone frozen and 30 epochs of fine-tuning with the last 30 layers selected while Batch Normalization layers remained frozen. Evaluation on 586 testing images achieved 95.56% accuracy, 98.79% Pneumonia precision, 95.09% recall, 96.90% F1-score, and 0.9929 ROC-AUC. These results indicate that the integrated preprocessing and Xception-based transfer learning framework provides strong and reproducible classification performance on the evaluated public dataset.
Unsupervised Anomaly Detection for Water Pump Fault Identification Under Extreme Class Imbalance Mochamad Rizal Fauzan; Ghayrie Habsy Alghifary; Muhamad Husen
Journal of Information System Exploration and Research Vol. 4 No. 4: October 2026
Publisher : shmpublisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joiser.v4i4.114

Abstract

Water pump fault detection in industrial systems is challenged by extreme class imbalance, where anomalous states account for less than 7% of operational data and labeled fault examples are scarce. This paper presents a comparative evaluation of three unsupervised anomaly detection approaches Isolation Forest (IF), Long Short-Term Memory Autoencoder (LSTM-AE), and One-Dimensional Convolutional Neural Network Autoencoder (1D-CNN-AE) applied to a real-world multivariate water pump sensor dataset comprising 220,320 one-minute observations across 51 sensors. All models are trained exclusively on normal operational data, eliminating dependence on fault labels. A sliding-window preprocessing pipeline (W = 30 timesteps) with a scaler fitted only on normal training data ensures no temporal data leakage. Anomaly thresholds are derived statistically (mean + 3σ) from reconstruction errors on the normal test partition. Using a chronological 60/40 train-test split and threshold-free metrics, Isolation Forest achieves AUROC = 0.9973 and Recall = 0.9850, while LSTM-AE yields F1 = 0.9218 and Precision = 0.8892. The results reveal a fundamental sensitivity-selectivity trade-off across unsupervised paradigms and provide actionable guidance for deploying anomaly detection in water infrastructure monitoring systems.

Page 1 of 1 | Total Record : 7

 1 

Filter by Year

2026 2026


Reset
Filter By Issues
All Issue Vol. 4 No. 4: October 2026 Vol. 4 No. 3: July 2026 Vol. 4 No. 2: April 2026