cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Esther Irawati Setiawan
Contact Email
esther@istts.ac.id
Phone
+62315027920
Journal Mail Official
insyst@istts.ac.id
Editorial Address
Kampus Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya) Ngagel Jaya Tengah 73-77, Surabaya, Indonesia
Location
Kota surabaya,
Jawa timur
INDONESIA
Insyst : Journal of Intelligent System and Computation
Published by Sekolah Tinggi Teknik Surabaya
ISSN : 26219220     EISSN : 27221962     DOI : https://doi.org/10.52985/insyst
Core Subject : Science,
Computer Science & IT
The Intelligent System and Computation Journal will be published for 2 editions in a year, every April and October. The Intelligent System and Computation Journal is an open access journal where full articles in this journal can be accessed openly. Review in this journal will be conducted with a blind review system. All articles in this journal will be indexed by Google Scholar. The topics contained in this journal consist of several fields (but not limited to): Algorithms and complexity Artificial Intelligence Big Data Analytics Biomedical Instrumentation Computational logic Computer Vision and Biometric Data and Web Mining Digital Signal Processing Image Processing Information Retrieval & Information Extraction Intelligence Embedded Systems Machine Learning Mathematics and models of computation Natural Language Processing Parallel & Distributed Computing Pattern Recognition Programming languages and semantics Speech Processing Virtual Reality & Augmented Reality
Arjuna Subject : Ilmu Komputer (semua kategori) - Kecerdasan Buatan
Articles 100 Documents
 6   7   8   9   10 
Comparison of Premium Rice Price Prediction in East Java with ARIMA and LSTM (Case Study: National Food Agency Data) Purwanto, Devi Dwi; Sitepu, Rasional; Honggara, Eric Sugiharto
Intelligent System and Computation Vol 6 No 2 (2024): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v6i2.407

Abstract

Rice price prediction plays a crucial role in maintaining economic stability and food security, especially in East Java, one of Indonesia's major rice production centers. This study aims to forecast premium rice prices in East Java using the ARIMA (AutoRegressive Integrated Moving Average) method. The data utilized in this research comprises premium rice prices obtained from the National Food Agency over the period from March 15, 2021, to October 17, 2024. The analysis process begins with data exploration to identify trends and seasonal patterns in the rice price data. Subsequently, the data is analyzed using ARIMA and LSTM methods, both recognized for their effectiveness in time-series forecasting. The ARIMA(1,1,1) model was selected due to its capability to capture price dynamics through its autoregressive, integrated, and moving average components, making it well-suited for linear data with minimal seasonal variation. LSTM was employed as a comparative model because it is a subset of Machine Learning that integrates computational models and neural network algorithms, offering potential improvements in prediction accuracy. The LSTM model used for prediction consists of four layers, each with 50 neurons, dropout rates of 20% and 30%, and a single output layer representing the predicted price. The results indicate the ARIMA model provides highly accurate price estimates with a Mean Absolute Percentage Error (MAPE) of 0.485%, whereas the LSTM model achieves a MAPE of 1.95%. These findings serve as a reference for policymakers and food industry stakeholders in formulating strategic measures to stabilize rice prices in East Java.
Comparison of Random Forest and SVM Algorithms in Credit Risk Evaluation Based on Debtor Occupation Prayesy, Putri Armilia; Pujakesuma, Angga
Intelligent System and Computation Vol 7 No 2 (2025): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v7i2.431

Abstract

Credit is one of the main sources of income for banking institutions and plays a crucial role in supporting long-term profit growth. However, credit distribution is inherently associated with risks, especially the risk of default when borrowers fail to meet their repayment obligations as agreed. One effective strategy to minimize such risks is to conduct a comprehensive and accurate creditworthiness assessment of prospective borrowers before loan approval is granted. This study aims to evaluate the performance of three classification algorithms—Random Forest, Support Vector Machine (SVM), and Artificial Neural Network (ANN)—in predicting credit risk based on the borrower’s occupation. The dataset used consists of 1,314 loan records with an imbalanced distribution between performing and non-performing loans. The experimental results show that the Random Forest algorithm achieved the highest accuracy at 97%, followed by Support Vector Machine at 73% and Artificial Neural Networks at 64%. While ANN is capable of capturing complex patterns through multilayered learning, Random Forest proved to be the most effective and robust in handling the given dataset. These findings clearly indicate that Random Forest can serve as a reliable method for financial institutions to enhance credit risk evaluation and minimize potential losses arising from loan defaults.
A Hierarchical Multi-Label Classification Approach for the Automated Interpretation of Spinal MRI Series Cahyadi, David; Pramana, Edwin; Limantara, Rudi; Wiguna, I Gusti Lanang Ngurah Agung Artha; Deslivia, Maria Florencia; Liando, Ivan Alexander
Intelligent System and Computation Vol 7 No 2 (2025): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v7i2.438

Abstract

Manually selecting MRI slices is a significant bottleneck in clinical workflows. This issue is worsened by inconsistent naming conventions and variable acquisition protocols across institutions and radiologists, often leading to redundant efforts and potential oversights during medical image data preprocessing. This study introduces a fully automated, four-level hierarchical classification system specifically designed to intelligently filter and select clinically relevant spinal MRI slices directly from raw DICOM series. Our primary objective is to streamline the initial stages of radiological assessment, ensuring that only pertinent images are presented for subsequent analysis and review. We thoroughly evaluated the performance of modern, efficient deep learning architectures, including EfficientViT, MobileNetV4, and RepViT, benchmarking them against a robust ResNet-18 baseline. The proposed pipeline systematically refines its analysis through a structured hierarchy: it first broadly identifies the anatomical region, then precisely classifies the spine location and specific view (axial, sagittal, or coronal). Subsequently, it categorizes the imaging contrast, and finally, confirms the presence of the spinal cord. Our comprehensive experimental results reveal that the EfficientViT-based model achieved the highest end-to-end F1-score of 0.8357, demonstrating robust accuracy across all classification levels. Furthermore, its average inference speed of 9.17 ms per image highlights its computational efficiency. This automated pipeline offers an effective and computationally efficient solution for speeding up initial medical image preprocessing, ensuring subsequent analytical tasks are performed on accurately selected, clinically relevant data.
Multi View Neural Network for Software Effort Estimation Prediction Setiawan, Boy; Subekti, Agus
Intelligent System and Computation Vol 7 No 2 (2025): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v7i2.442

Abstract

Software Effort Estimation (SEE) is a critical challenge in software project management, dating back to the early years of software engineering. Accurate estimation of the effort required for software development is essential for project planning, resource allocation, and risk management. Incorrect effort estimates can result in poor resource distribution, cost overruns, missed deadlines, and even complete project failure. This issue is increasingly urgent today as software systems are deeply embedded in almost every product and service, amplifying the need for reliable and accurate predictions. Over the years, several methods for SEE have been proposed, ranging from algorithmic models to expert judgment. More recently, machine learning (ML) approaches such as Case-Based Reasoning (CBR), Support Vector Machines (SVM), Decision Trees (DT), and Neural Networks (NN) have gained attention for their ability to model complex, nonlinear relationships inherent in SEE tasks. In this study, we propose a novel approach based on multi-view learning with NN (MVNN), which leverages multiple views from existing datasets, thus improving performance and generalization, particularly when the available data is small and scarce. The effectiveness of the MVNN model is validated through empirical comparisons with existing SEE models, demonstrating its potential to enhance SEE accuracy and improve prediction reliability.
Implementation of Reinforcement Learning Agent Plugins Using A2C and PPO Methods in the Godot Engine Paulus Lucky Tirma Irawan; Marvin Adinata; Mochamad Subianto
Intelligent System and Computation Vol 8 No 1 (2026): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v8i1.470

Abstract

Reinforcement learning (RL) has become one of the most dynamic areas of artificial intelligence, enabling agents to learn optimal behaviour through trial-and-error interactions with their environment. Most RL studies and implementations are conducted using Python-based libraries or within the Unity engine. However, the open-source Godot Engine, which has gained significant traction among indie developers, offers an attractive alternative platform for RL research and interactive game development. This study presents the implementation of RL agents using the Advantage Actor-Critic (A2C) and Proximal Policy Optimization (PPO) algorithms on the Godot Engine through the AgentRL plugin. Two classic RL environments, Cart Pole and Cliff Walking, were replicated in Godot to evaluate algorithm performance. The AgentRL plugin facilitates real-time communication between Godot and Python, allowing model training through the Stable-Baselines3 library. Performance was measured using average episode length and average reward, and statistical analysis was conducted using Welch’s t-test. Experimental results indicate that PPO outperformed A2C in both environments: in Cart Pole, PPO achieved an average of 115.93 steps longer balance duration, while in Cliff Walking, it produced 6.19 steps shorter episodes and 27.93 points higher rewards. These findings confirm that PPO offers better training stability and efficiency than A2C. Furthermore, the results demonstrate that Godot, integrated with AgentRL, is a viable and flexible platform for reinforcement learning research and can serve as a foundation for future studies in AI-driven game development.
Detecting Lung Disease Based on Chest X-ray Images Using a Hybrid CNN-KELM Approach Dian Candra Rini Novitasari; Musfiroh Musfiroh; Dina Zatusiva Haq
Intelligent System and Computation Vol 8 No 1 (2026): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v8i1.479

Abstract

Tuberculosis (TB) is a disease caused by the Mycobacterium tuberculosis (M.tb) bacterium. TB ranks among the top 10 deadliest diseases worldwide and is the second most contagious disease after COVID-19. The World Health Organization (WHO) recommends using Chest X-ray (CXR) imaging techniques, given their high sensitivity and cost-effectiveness. This study proposes a hybrid CNN-KELM (CKELM) method for the classification of four lung disease categories based on chest X-ray (CXR) images: tuberculosis, pneumonia, COVID-19, and normal, all within a short computational time. This study experimented with several types of CNN architectures implemented for feature extraction, while KELM for classification used hyperparameters that tested various kernel types and regularization coefficients. The experimental results indicate that the best performance is achieved using the DenseNet201 architecture with a polynomial kernel and a regularization coefficient of 0.1. The polynomial kernel demonstrates superior performance across all CNN architectures. Furthermore, a regularization coefficient of 0.1 exhibits the highest accuracy in the kernel and CNN architecture experiments. The DenseNet201-KELM model attains an accuracy, sensitivity, specificity, precision, and F1-score of 99.57%, 99.57%, 99.86%, 99.57%, and 99.57%, which is 7% better than without under sampling and detection using the DenseNet201-KELM method requires a computational time of 309.19 seconds. The proposed method achieved good performance in multi-class classification, especially for balanced data, with fast computational time.
An Explainable Hierarchical Rule-Based Model for Inventory Optimization in Data-Scarce Culinary MSMEs Yaslinda Lizar; Asriwan Guci
Intelligent System and Computation Vol 8 No 1 (2026): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v8i1.480

Abstract

Inventory management is a critical component of supply chain operations, particularly for culinary Micro, Small, and Medium Enterprises (MSMEs) that face volatile demand patterns and perishable raw materials. Inaccurate procurement decisions may result in overstocking, stock shortages, and financial losses. Although artificial intelligence approaches such as machine learning have been widely applied in inventory forecasting, these methods typically require large-scale historical datasets and advanced computational infrastructure, which are often unavailable in MSME environments. This study proposes an explainable hierarchical rule-based inference model designed to support inventory optimization in data-scarce culinary MSMEs. The model integrates dynamic operational factors, including day type, weather conditions, supplier lead time, and special events, into a transparent decision-making mechanism based on IF–THEN rules. The research adopts a Research and Development methodology using the Waterfall framework, covering requirement analysis, rule-base construction, system implementation, and evaluation. Model validation was conducted through 20 expert-verified decision scenarios and assessed using confusion matrix metrics. The experimental results demonstrate that the proposed system achieved 90% accuracy, 92.3% precision, 92.3% recall, and a 92.3% F1-score when compared with expert procurement decisions. These findings indicate that explainable rule-based systems remain a practical and reliable solution for inventory decision support in culinary MSMEs with limited data resources.
Optimizing KNN: Impact of Distance Metrics and SMOTE on Heart Disease Classification Windra Swastika
Intelligent System and Computation Vol 8 No 1 (2026): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v8i1.481

Abstract

Heart disease remains the leading cause of mortality worldwide, accounting for approximately 32% of all global deaths. The development of accurate and clinically reliable machine learning-based prediction systems is therefore essential for supporting early clinical decision-making. This study proposes a comprehensive optimization framework for the K-Nearest Neighbor (KNN) algorithm applied to heart disease classification using the UCI Cleveland Heart Disease Dataset. While prior work has addressed class imbalance using the Synthetic Minority Over-sampling Technique (SMOTE) and feature scaling via Min-Max Normalization, no study has simultaneously investigated the effect of distance metric selection and systematic K value optimization in the context of preprocessed imbalanced medical data. This paper makes three contributions: (1) a comparative analysis of three distance metrics, Euclidean, Manhattan, and Minkowski (p=3), applied to KNN after preprocessing; (2) systematic optimal-K identification using Grid Search with Stratified 10-Fold Cross-Validation across all metric-scenario combinations; and (3) a structured ablation study across four preprocessing scenarios to quantify the individual and combined contributions of SMOTE and Min-Max Normalization. Experiments were conducted on 297 samples with 13 clinical features. Results show that the best clinically oriented model (Scenario C: SMOTE + Manhattan, K=9) achieves 81.67% accuracy, 82.14% recall, and 80.70% F1-score. The Minkowski metric in the fully combined scenario (D) achieves the highest AUC of 92.47%, with optimal K=21, a markedly different configuration than Euclidean and Manhattan, which converge at K=1. These findings demonstrate that distance metric choice and K optimization interact significantly, offering practical configuration guidelines for KNN in medical classification tasks.
Loyal Customer Segmentation Using RFM Model and K-Means Algorithm on E-Commerce Data Hersa Safitri; Haviluddin Haviluddin; Anton Prafanto
Intelligent System and Computation Vol 8 No 1 (2026): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v8i1.495

Abstract

Comprehensive customer segmentation in the e-commerce context is essential for supporting effective data-driven decision-making. This research aims to develop and validate a loyal customer segmentation model by integrating Recency, Frequency, and Monetary (RFM) analysis with the K-Means clustering algorithm to generate stable, well-separated, and managerially relevant clusters. The dataset analyzed comprises 392,732 transactions from 4,339 customers, obtained from a public e-commerce platform. The research workflow includes data preprocessing, RFM score computation, feature standardization, determination of the optimal number of clusters using the Elbow Method, and cluster evaluation using the Silhouette Coefficient and Davies–Bouldin Index (DBI). The experimental results indicate that a five-cluster configuration yields the best performance, achieving a Silhouette score of 0.6158 and a DBI of 0.7190. The stability of the five-cluster solution is confirmed over 30 random initializations (Silhouette = 0.6159 ± 0.0031; mean Adjusted Rand Index = 0.9892), and its practical value is demonstrated through a segment-level revenue contribution analysis and a comparison against six baseline methods. The Champions segment (comprising Top Champions and Champions) accounts for only 0.3% of the customer base yet contributes the highest total monetary value of US$190,808.54. In contrast, 94.5% of customers are classified as at risk or hibernating. These findings demonstrate that integrating RFM with K-Means, validated across multiple evaluation metrics, yields a reliable, measurable, and actionable customer segmentation framework. This approach effectively supports the development of customer retention strategies and the optimization of data-driven customer value.
Experimental Web Server Incident Detection Based on Wazuh-Snort using SPDLC Method Agus Wijayanto; Ahmad Yazid Bustomi; Syaddam; Tri Sudinugraha
Intelligent System and Computation Vol 7 No 2 (2025): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v7i2.477

Abstract

Web servers are essential but vulnerable to network attacks such as SYN Flood and port scanning. This study designs and implements an incident detection system by integrating Wazuh, Snort (IDS), and Telegram notifications following the Security Policy Development Life Cycle. Snort functions as a network sensor generating alerts forwarded to Wazuh for centralized log management and correlation; critical alerts are pushed to administrators via a Telegram bot for real-time response. The experimental environment uses virtualized machines that host a Wazuh server, a Snort IDS, a web server, and an attacker node. Evaluation metrics include detection accuracy, false-negative rate, and notification response time under two attack scenarios: SYN Flood and port scanning. Results indicate detection accuracy of 41.34% and a false-negative rate of 58.6% for SYN Flood attacks, with Snort-to-Wazuh latency of 13.559 ms and Telegram delivery of 0.4 s. Port scanning was detected with 100% accuracy and 0% false-negative rate, recording Snort-to-Wazuh latency of 1.490 ms and Telegram delivery of 1.33 s. The integration enhances centralized visibility and accelerates administrator awareness, yet it exhibits limitations for high-volume DoS traffic due to log buffering and throughput constraints. Recommendations include increasing Wazuh agent buffer capacity, optimizing Snort rules, and deploying higher-capacity hardware to improve detection under heavy attack loads.

Page 10 of 10 | Total Record : 100

 6   7   8   9   10 

Filter by Year

2019 2026


Reset
Filter By Issues
All Issue Vol 8 No 1 (2026): INSYST: Journal of Intelligent System and Computation Vol 7 No 2 (2025): INSYST: Journal of Intelligent System and Computation Vol 7 No 1 (2025): INSYST: Journal of Intelligent System and Computation Vol 6 No 2 (2024): INSYST: Journal of Intelligent System and Computation Vol 6 No 1 (2024): INSYST: Journal of Intelligent System and Computation Vol 5 No 2 (2023): INSYST: Journal of Intelligent System and Computation Vol 5 No 1 (2023): INSYST: Journal of Intelligent System and Computation Vol 4 No 2 (2022): INSYST: Journal of Intelligent System and Computation Vol 4 No 1 (2022): INSYST: Journal of Intelligent System and Computation Vol 3 No 2 (2021): INSYST: Journal of Intelligent System and Computation Vol 3 No 1 (2021): INSYST: Journal of Intelligent System and Computation Vol 2 No 2 (2020): INSYST:Journal of Intelligent System and Computation Vol 2 No 1 (2020): INSYST:Journal of Intelligent System and Computation Vol 1 No 2 (2019): INSYST:Journal of Intelligent System and Computation Vol 1 No 1 (2019): INSYST: Journal of Intelligent System and Computation More Issue