cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Christian Harito
Contact Email
christian.harito@binus.edu
Phone
+6221-5350660
Journal Mail Official
aagung@binus.edu
Editorial Address
Universitas Bina Nusantara Jl. Kebon Jeruk Raya No.27 Kebon Jeruk, Jakarta Barat 11530
Location
Kota adm. jakarta barat,
Dki jakarta
INDONESIA
Engineering, Mathematics and Computer Science Journal (EMACS)
Published by Universitas Bina Nusantara
ISSN : -     EISSN : 26862573     DOI : https://doi.org/10.21512/emacs
Core Subject : Science, Education, Engineering,
Civil Engineering, Building, Construction & Architecture Computer Science & IT Engineering Industrial & Manufacturing Engineering Mathematics
Engineering, MAthematics and Computer Science (EMACS) Journal invites academicians and professionals to write their ideas, concepts, new theories, or science development in the field of Information Systems, Architecture, Civil Engineering, Computer Engineering, Industrial Engineering, Food Technology, Computer Science, Mathematics, and Statistics through this scientific journal.
Arjuna Subject : Teknik (semua kategori) - Teknik (Lain-Lain)
Articles 174 Documents
 14   15   16   17   18 
Gender Classification Using Keystroke Dynamics: Enhancing Performance with Feature Selection and Random Forest Maulina, Ayu; Charisma, Rifqi Alfinnur
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.13445

Abstract

The purpose of this study is to improve gender categorization by examining the usage of keyboard dynamics, with enhanced model performance through data standardization and appropriate feature selection. Features including gender, age, handedness, language, education, and metrics measuring typing behavior like mean_latency, std_latency, and frequency are all included in the dataset. Correlation analysis served as the foundation for the feature selection procedure, which is essential for effective model training, and data normalization was performed to guarantee consistency among the characteristics that were chosen. Because of its stability and capacity to handle complicated data, the Random Forest classifier was selected. The findings demonstrate that the Random Forest model achieved an accuracy of 95% and an F1-score of 95% when using all features, and 82% accuracy with an F1-score of 82.5% when using only the selected features. The results emphasize how important it is to choose the appropriate characteristics and standardize the data in order to increase predictive accuracy. By showcasing keystroke dynamics' capacity for gender categorization, this study advances the area and creates opportunities for further research in user experience improvement, digital service customization, and online behavioral analysis. Overall, the study emphasizes the importance of feature engineering, normalization, and model tuning for achieving accurate and reliable classification outcomes.
Cost Analysis of Construction Cost Planning for Landfill Site Junaedi, Nurhayati; Bayuaji, Ridho; Susilo, Alfred Jonathan
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.13800

Abstract

Accurate construction cost planning is essential to ensure project success, as inaccurate estimates may lead to delays, cost overruns, and reduced quality. Landfill construction, which is more complex than many other infrastructure projects, includes several components such as landfill work, leachate treatment facilities, and supporting infrastructure. The purpose of this study is to determine the major cost factors that have a major impact on the overall cost of building landfill sites.  With the aid of SPSS software, a regression analysis was carried out using cost data from six landfill projects in Java Island that were completed between 2013 and 2018.  With a Sig value of 0.000 (<0.05) and a very strong correlation (Pearson Correlation 0.991, within the 0.8–1.0 interval), the results show that landfill work (X3) significantly affects total costs (Y).  Leachate treatment facilities (X4) are another crucial element in a number of situations, but landfill work (X3) consistently represents the largest portion of construction costs, according to proportion analysis. These results demonstrate the growing significance of environmental facilities and point to landfill work as the main factor influencing landfill construction costs.  The study offers contractors and federal and local governments useful information for creating more precise cost estimates, maximizing budgetary allotments, and enhancing planning and development efficiency for landfill projects.
Rice Hull Management System: A Mobile-based Application Tool for Cooperatives Usita, Maricris; Timalog, Cris Ann Fogusa; Javier, Maychiel; Indap, Jhune Carlo; Calera, Gricelyn; Ramos, Jessa Jane
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.13839

Abstract

This study focuses on the design and thorough assessment of the Rice Hull Management System RHMS for mobile devices, which targets cooperatives located specifically in Occidental Mindoro, Philippines. The Philippines is well-known for its rice production; however, there are severe concerns regarding environmental sustainability due to the poor management of rice hulls. To offer a solution, the Rice Hull Management System mobile application was created for user cooperatives from Occidental Mindoro. The application was developed using the RAD methodology with React Native and Firebase, enabling the system to be responsive, scalable, and secure. Through the utilization of RHMS, rice hull transactions are processed more efficiently with automated summaries, precision reports, advanced analytics, and real-time updates, all of which facilitate information-based decision-making and foster eco-friendly agricultural practices within and outside the region. The implementation evaluation of the system was conducted using a combination of surveys, usability tests, and performance benchmarks, which included IT specialists, cooperative staff, and members. System reliability was demonstrated to be high, with a Cronbach’s alpha greater than 0.80 and high user satisfaction, with grand mean scores ranging from 4.08 to 4.23 (“Very Good”). Evaluated criteria for the RHMS included efficiency, integration, usability, reliability, safety, and mobility, all of which received excellent ratings, confirming the system's effectiveness in resolving operational manual inefficiencies and enhancing transparency. This study focuses on the application of technology in rice hull waste management to promote environmental sustainability while meeting the requirements of agricultural cooperatives. The RHMS showcases considerable promise for development and implementation across agricultural supply chains, given its secure, easy-to-use, and flexible interface for users, administrators, and cooperatives. Proposed recommendations included continuous system enhancements, compliance with health regulations, integration with other platforms, and training programs to foster sustained system utilization and impact.
Comparative Analysis of CNN, LSTM, and CNN–LSTM for Indonesian Stock Prediction Joddy, Setiawan
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.14326

Abstract

Predicting the stock market remains a challenging task brought by the nonlinear, volatile, and dynamic nature of financial time series. While deep learning techniques have been widely applied in developed markets, studies in emerging markets such as Indonesia remain scarce. This study conducts a comparative analysis of three deep learning models—Convolutional Neural Networks (CNN), Long Short-Term Memory networks (LSTM), and a hybrid CNN–LSTM—on five randomly selected constituents of the IDX30 index. The data range from January 2020 to December 2024, providing a general view of stock movement in recent years. The models were trained on daily OHLCV (Open, High, Low, Close, Volume) data, which was formatted using a sliding-window approach. Results show that LSTM achieved the lowest RMSE of 0.0222 ± 0.0030, MAE of 0.0172 ± 0.0015, and the highest R² of 0.889 ± 0.068. The Hybrid model delivered intermediate performance, improving upon CNN but not surpassing LSTM. These findings confirm that LSTM networks are particularly effective for stock price forecasting in the Indonesian market, while hybrid CNN–LSTM architectures can provide complementary strengths by balancing short-term feature learning with long-term temporal dependencies.
Integration of Multi-Architecture Deep Learning Models for Pneumonia Detection Based on Chest X-Ray Imaging Edbert, Ivan Sebastian; Oktovianus, Louis; Tanriwan, Robert; Aulia, Alvina
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.14336

Abstract

Pneumonia remains a leading cause of child mortality worldwide, particularly in resource-limited settings where diagnostic tools and expertise are scarce. Recent advances in deep learning offer an opportunity to enhance pneumonia detection through automated analysis of chest X-ray images. This study evaluates the performance of ten state-of-the-art deep learning architectures, including VGG16, ResNet50, DenseNet121, and MobileNetV2, for pneumonia detection using the widely recognized "Chest X-Ray Images (Pneumonia)" dataset. The dataset underwent rigorous preprocessing, including image resizing, data augmentation, and class balancing, to optimize model training and improve generalization. Performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC were utilized to assess model effectiveness. Among the evaluated architectures, MobileNetV2 demonstrated the best performance with an accuracy of 97.51% and an AUC of 0.9941, highlighting its potential for reliable diagnostic applications. The results also emphasize the trade-offs between sensitivity and specificity across models, offering useful insights for real-world deployment. This study underscores the importance of leveraging deep learning models in clinical diagnostics, particularly in environments with limited healthcare resources. Beyond evaluating models, the findings provide evidence-based recommendations for selecting efficient architectures that balance accuracy and computational efficiency. Future work will focus on integrating multimodal datasets, improving explainability, and validating these models in diverse clinical environments to ensure scalability, trust, and generalizability for global health applications.
Comparison of IndoBERT and SVM Algorithm to Perform Aspect Based Sentiment Analysis using Hierarchical Dirichlet Process Octarini, Sheila Prima; Zakiyyah, Alfi Yusrotis; Purwandari, Kartika
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.13493

Abstract

Analyzing the performance of SVM and IndoBERT models for aspect-based sentiment analysis on fashion reviews in Tokopedia E-Commerce. This study employs the SMOTE technique due to the imbalance in the original data. Aspect determination using the Hierarchical Dirichlet Process (HDP) model yields satisfactory results with an adequate coherence score. The comparison between SVM and IndoBERT methods for aspect-based sentiment analysis shows that SVM is superior. IndoBERT achieved an accuracy of 87%, precision of 91%, recall of 93%, and F1-Score of 92%, while SVM attained an accuracy of 96%, precision of 100%, recall of 92%, and F1- Score of 96%. Therefore, the SVM model was chosen for implementation on a website that allows users to view aspect-based sentiment analysis on products in E-Commerce. The HDP model effectively grouped related terms into aspects such as “Material,” “Shipping,” and “Colour,” enhancing interpretability in sentiment classification. The resulting website enables users to analyze product sentiments interactively, providing actionable insights for both sellers and customers to assess product quality and service satisfaction more efficiently.
Low-Resolution Face Recognition: A Review of Methods and Data Utomo, Yesun; Kelvin
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.14082

Abstract

This paper provides a review of previous studies in Low-Resolution Face Recognition (LRFR), specifically focusing on cross-resolution Face Recognition (FR) methods. While state-of-the-art deep learning FR systems achieve high accuracy on high-resolution (HR) images, they are generally unsuitable for low-resolution (LR) images frequently encountered in applications like surveillance systems, where faces often have low pixel counts due to capture conditions. Cross-resolution FR, which compares an HR image with an LR image, presents a significant challenge due to the distinct visual properties of images at different resolutions. The paper discusses two primary approaches to address this problem: super-resolution (SR), which is a transformative method that aims to construct HR images from LR ones, and unified feature space (UFS), a non-transformative method that maps facial features from varying resolutions into a shared feature space. This work summarizes both SR and UFS methods. Based on the review, the paper concludes that non-transformative (UFS) methods are more suitable for future directions. This recommendation is driven by their lower computational power requirements, proven effectiveness in real-world implementations such as mobile devices and drones, and alignment with current technological trends. The paper also emphasizes the need for further research using real or natural LR face images to identify degradation patterns and compare results between real and artificially generated LR images.
Enhancing Tourism Demand Forecasting Accuracy Through Clustering Time Series: A Comparison MAPE Analysis of Indonesian Provincial Domestic Tourist Flows Purnama, Mohammad Dian
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.14112

Abstract

The post-pandemic recovery period of the Indonesian tourism sector poses new challenges for accurate tourism demand forecasting across Indonesia's diverse provincial richness. This research aims to enhance the predictive accuracy of domestic tourism demand by comparing conventional single-provincial forecasting methods with clustering-based time series techniques. The Geometric Brownian Motion (GBM) model analyzed data regarding the monthly influx of domestic tourists to 34 provinces from January 2021 to May 2025. This study utilized average linkage agglomerative nesting (AGNES) clustering to discern structural similarities among provinces. Subsequently, silhouette analysis was employed to determine the optimal number of clusters. The findings demonstrate that the cluster-based forecasting approach markedly improved accuracy relative to the non-clustered model. The Mean Absolute Percentage Error (MAPE) for the traditional provincial forecasts was 16.48%. The first cluster-based model had an MAPE of 13.38% and the second cluster-based model had an MAPE of 6.54%. These findings indicate that grouping provinces with analogous temporal patterns enhances the model's ability to identify the underlying dynamics in domestic tourism flows. The work underscores the efficacy of combining stochastic models with hierarchical clustering to enhance evidence-based tourist planning and policy development. This study improves sustainable tourism management by providing an empirical foundation for enhanced forecasting precision, particularly in post-crisis recovery periods.
Comparative Study of CNN-based Deep Learning Models for Animal, Digit, and Flower Image Classification Suri, Puti Andam; Setiono, Michael Alvin; Andrew, Andrew; Fajar, Muhammad
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.14317

Abstract

This study explores how four convolutional neural network (CNN) models MobileNetV2, DenseNet121, EfficientNetB0, and InceptionV3 perform in classifying images from three different datasets: animals, handwritten digits (MNIST), and flowers. The main goal is to understand which model offers the best balance between accuracy and efficiency when applied to datasets with varying complexity. Each model was trained and tested using identical preprocessing steps, and its performance was evaluated based on accuracy, precision, recall, and F1-score through a confusion matrix. Training and testing times were also measured to assess computational efficiency. The results show that DenseNet121 consistently achieved the highest accuracy: 98% on animal images and 88% on flower images, while MobileNetV2 provided a close performance (97% and 82%) but with much faster processing times, between 11 and 55 minutes. EfficientNetB0, on the other hand, performed poorly on the more complex flower dataset, achieving only 5% accuracy. These findings suggest that DenseNet121 is ideal for projects where accuracy is the main concern, whereas MobileNetV2 is more suitable for real-time applications that require quick responses without a major drop in accuracy. Overall, this research highlights the importance of aligning model selection with both dataset characteristics and computational limitations in practical image classification tasks.
Optimizing Enterprise Risk Management for Decision Making Using Knowledge Graph Albone, Aan
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 3 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i3.14325

Abstract

The challenge in current enterprise risk management is that hundreds of risks are eventually recorded without knowing how hazards relate to one another or cascade. The distinction between peripheral and critical hazards is unknown to decision-makers. Organizations can depict the interconnectedness of risk in a structured, adaptable, and understandable way by showing these components as nodes and their interactions as edges. This knowledge graph makes it possible to store and query risk data in ways that are not entirely supported by conventional relational models. This method's ability to execute graph queries that uncover links and patterns that would otherwise be obscured in siloed datasets is one of its main advantages. Such inquiries can reveal how a single threat can lead to many vulnerabilities across multiple assets, or how flaws in shared systems can directly and indirectly raise exposure to interconnected hazards. These revelations draw attention to structural flaws that linear or isolated investigations frequently ignore. Organizations can improve situational awareness and long-term risk governance by using such a knowledge graph to find hidden trends, pinpoint important risk spots, and more efficiently prioritize mitigation efforts. The knowledge graph also helps to optimize enterprise risk management goals like resource allocation, control prioritization, and prompt reaction planning. Enterprise risk management can effectively represent the intricate relationships between risks, vulnerabilities, threats, and assets by incorporating a knowledge graph. Businesses can concentrate mitigation efforts where they will have the biggest impact by determining which nodes and edges are the most important and highest impact. This focused strategy increases overall resilience and decreases inefficiencies.

Page 17 of 18 | Total Record : 174

 14   15   16   17   18 

Filter by Year

2019 2026


Reset
Filter By Issues
All Issue Vol. 8 No. 1 (2026): EMACS (In Press) Vol. 7 No. 3 (2025): EMACS Vol. 7 No. 2 (2025): EMACS Vol. 7 No. 1 (2025): EMACS Vol. 6 No. 3 (2024): EMACS Vol. 6 No. 2 (2024): EMACS Vol. 6 No. 1 (2024): EMACS Vol. 5 No. 3 (2023): EMACS Vol. 5 No. 2 (2023): EMACS Vol. 5 No. 1 (2023): EMACS Vol. 4 No. 3 (2022): EMACS Vol. 4 No. 2 (2022): EMACS Vol. 4 No. 1 (2022): EMACS Vol. 3 No. 3 (2021): EMACS Vol. 3 No. 2 (2021): EMACS Vol. 3 No. 1 (2021): EMACS Vol. 2 No. 3 (2020): EMACS Vol. 2 No. 2 (2020): EMACS Vol. 2 No. 1 (2020): EMACS Vol. 1 No. 1 (2019): EMACS More Issue