Jurnal ICT : Information and Communication Technologies

Application of the Analytical Hierarchy Process and GIS in a Decision Support System for Determining the Location of the Final Disposal Site (FDS) for the City of Medan in Deli Serdang Regency

2026-05-13T07:29:19+00:00

Indonesian Sign Language (BISINDO) is the primary communication medium for the deaf community, yet limited public understanding often leads to communication barriers. Previous sign language recognition studies have generally been conducted offline, lacked real-time web integration, and produced only text-based outputs without multimodal interaction. To address these limitations, this study proposes a real-time web-based BISINDO translator system using a hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model, integrated with a Text-to-Speech (TTS) feature. The novelty of this research lies in the combination of CNN for spatial feature extraction and LSTM for temporal sequence learning within a fully deployed web application framework (NzSignify), enabling real-time end-to-end sign language translation with both text and voice output. The dataset consists of primary video recordings from three subjects, covering 11 gesture classes with 1,000 grayscale frames per class at a resolution of 100×89 pixels. The proposed model is implemented using a React.js and Node.js-based system to support real-time inference. Experimental results show that the hybrid CNN-LSTM model achieves a classification accuracy of 96% based on Confusion Matrix evaluation. In real-time testing, an 80% confidence threshold effectively filters misclassified gestures and improves translation reliability into text and speech outputs. Compared to previous studies that mainly rely on standalone CNN or traditional machine learning methods with offline processing, the proposed approach demonstrates improved capability in capturing both spatial and temporal features of sign gestures as well as supporting real-time deployment. These findings indicate that the developed system provides a more practical, accurate, and interactive solution for BISINDO translation, enhancing communication accessibility between deaf and hearing communities through a real-time multimodal platform.

An Evaluation Of Project Manager Performance In The Success Of Agile Scrum-Based Application Development Projects

2026-05-07T03:36:34+00:00

The role of a Project Manager (PM) is vital to the success of application development projects. Within the Agile Scrum framework, a PM ensures effective team collaboration, adaptability to changing requirements, and the timely completion of sprints. This study aims to evaluate Project Managers' performance in relation to the success of Agile Scrum-based application development projects. Using a quantitative approach a survey was conducted with 20 respondents from Agile-based software development teams. The study reveals that leadership, communication skills, and decision-making are the most significant contributors to project success. These factors collectively account for 78% (R² = 0.78) of the project success rate reported. The TELOS feasibility analysis indicates high feasibility in the technical (90%), operational (85%), and schedule (80%) dimensions, while the economic (75%) and legal (70%) dimensions identified areas for improvement. Ultimately, the performance of the Project Manager plays a dominant role in determining the success of Agile Scrum project implementation.

Design of an IoT-Based Earthquake Vibration Detection System Using MPU6050 Sensors with Real-Time Monitoring via a Mobile App

2026-05-13T07:26:34+00:00

Advances in Internet of Things (IoT) technology enable electronic devices to communicate and exchange data in real time. This study aims to design and implement an IoT-based earthquake vibration detection system using an MPU6050 sensor and an ESP32 microcontroller with real-time monitoring via a mobile application. The research method used is engineering research with an iterative prototype approach. The system consists of an MPU6050 sensor as a vibration detector, ESP32 as a data processor and communication module, an IoT server, an LCD display, a buzzer, and a mobile application as the user interface. Acceleration data on the X, Y, and Z axes is processed using a threshold-based method to distinguish between normal conditions and earthquake events. The test results show that the system achieves a detection accuracy of 92.6%, with an average response time of 1.2 seconds from vibration detection to mobile notification delivery. In addition, the system demonstrates a data transmission reliability of 98.3%, indicating stable communication between the device and the IoT server. The system is capable of detecting vibration changes effectively, transmitting data in real time, and displaying monitoring information through a mobile application. Early warning notifications are successfully generated when vibration values exceed the defined threshold. Based on the results, the proposed system provides a low-cost, efficient, and easy-to-implement solution for earthquake vibration monitoring on a local scale. However, improvements are still required in adaptive threshold optimization and large-scale field testing to enhance system robustness and reliability under real-world conditions.

An Expert System for Diagnosing Laptop Hardware Failures Using a Web-Based Certainty Factor Method

2026-05-07T03:19:07+00:00

Rapid technological developments are driving the increased use of laptops in various activities, which indirectly increases the potential for hardware damage. In laptop repair services, the damage identification process generally relies on the technician's experience, resulting in subjective and potentially inconsistent decisions. This situation demonstrates the need for a system capable of providing structured and measurable diagnostic support. The approach used is the Certainty Factor method to represent the level of confidence in diagnostic results. The developed web-based expert system allows users to identify damage based on selected symptoms, then produces output in the form of the type of damage along with a confidence value as a percentage. Evaluation is carried out by comparing the system's diagnostic results with those of experts in a number of test cases. The test results show a good level of agreement, so the system can be used as an aid in the initial diagnosis of laptop hardware damage.

Analysis of Public Sentiment Toward Mental Health on Social Media Using Naïve Bayes

2026-05-19T15:03:43+00:00

Mental health is a global issue that has garnered significant public attention on social media. The platform X (formerly Twitter) is widely used by the public to openly express emotional conditions, yielding vast amounts of unstructured textual data. This research aims to analyze public sentiment regarding mental health issues on social media X using the Multinomial Naïve Bayes algorithm combined with Term Frequency-Inverse Document Frequency (TF-IDF) word weighting. The dataset consists of 9,000 tweets written in Indonesian, collected between February 15 and 27, 2025, using the keywords kesehatan_mental (mental health), stress (stress), kecemasan (anxiety), and depresi (depression). To enhance data quality, a comprehensive text preprocessing pipeline was implemented, including cleaning, case folding, word normalization (using a 59-entry mapping dictionary), tokenizing, stopword removal, and stemming. The performance of the classification model was evaluated using a confusion matrix on 1,800 test data. The results demonstrate that the Multinomial Naïve Bayes model achieved a high accuracy of 90.78% and a macro average F1-score of 90.75%. Specifically, the positive sentiment class yielded a precision of 96.22% and a recall of 84.89%, while the negative sentiment class achieved a precision of 86.48% and a recall of 96.67%. Furthermore, this study integrates the classification model into a web-based system equipped with an explainability feature that visualizes word contributions to the sentiment outcomes. This research contributes an interpretative, informative, and efficient computational approach for monitoring public sentiment trends toward mental health issues on Indonesian social media.

Metadata and Location-Based Digital Document Authenticity Identification

2026-05-07T02:11:49+00:00

The growth of digital documents has simplified the creation, storage, and distribution of information, but it has also increased the possibility of document forgery and manipulation. This study implements a digital forensic approach to support the identification of digital document authenticity through metadata and location analysis. The research method follows the National Institute of Standards and Technology (NIST) framework, consisting of collection, examination, analysis, and reporting. The system was developed as a web-based application using HTML, CSS, and JavaScript, with support for JPG, PNG, PDF, and DOCX files. The analyzed metadata includes file type, file size, creation time, modification time, software or device information, author identity, and GPS coordinates when available. The system was tested using black box testing on key features, including file upload, fetch URL, export JSON, clear, and GPS location display. The results show that the system is able to extract and present metadata in a structured manner and support the initial verification of digital document authenticity. Metadata can be used as an early indicator to detect inconsistencies in document history, although the reliability of the analysis depends on the completeness of metadata stored in the examined file.

Application of The Random Forest Algorithm in Classifying the Tendency of Impulsive Purchasing Behavior Among Gen Z Consumers in E-Commerce Based on Flash Sale Features

2026-05-13T07:34:46+00:00

The rapid growth of e-commerce in Indonesia, particularly on Shopee, has significantly influenced consumer behavior through promotional strategies such as flash sales. This study aims to classify impulsive buying tendencies among Generation Z, identify key influencing factors, and develop a web-based classification system for behavioral analysis. A quantitative data mining approach was applied using the Random Forest algorithm. The dataset consisted of 420 Gen Z respondents collected through a Likert-scale questionnaire using purposive sampling, and model evaluation was conducted using 10-fold cross-validation to ensure reliability. The results show that the Random Forest model achieved an accuracy of 83.16%, outperforming Decision Tree (78.42%) and Logistic Regression (75.08%), indicating its effectiveness in handling complex behavioral patterns. Feature importance analysis revealed that limited stock availability (39.85%) and discount magnitude (33.21%) are the most dominant factors influencing impulsive buying behavior, followed by promotional duration and notification attractiveness. These findings emphasize the role of urgency and scarcity in driving impulsive purchases among Gen Z consumers. Additionally, a web-based system was developed using the Flask framework in Python to support automated data processing, model training, and visualization of results. The system enables real-time behavioral analysis and decision support for digital marketing strategies. Overall, the study demonstrates that machine learning, particularly Random Forest, provides a more accurate and objective approach for analyzing impulsive buying behavior compared to conventional statistical methods, while also offering a practical tool for e-commerce analytics and strategy optimization.

Development of a Real-Time Web-Based Battalion Picket Activity Monitoring System: A Case Study of the INFO GIAT Application

2026-05-07T03:45:28+00:00

Operational effectiveness in a Battalion environment depends heavily on the discipline and orderly execution of picket duties. However, reliance on manual administration such as physical journals and fragmented data recapitulation creates bottlenecks in the flow of vital information, increases the risk of data loss, and limits leadership’s ability to supervise unit activities in real-time. This research develops INFO GIAT (Informasi Gerakan Internal Anggota Terpadu), a web-based information system designed to transform battalion picket management into an integrated digital ecosystem. The Prototyping development method was adopted to build the system iteratively and responsively to user needs, while Blackbox Testing was used to validate system reliability. The system incorporates Role-Based Access Control (RBAC) to manage access for three user types: Administrator, Organik/Atasan (supervisor), and Kadet (picket officer). A key security feature is a two-factor authentication (2FA) mechanism via Telegram API. Results indicate that INFO GIAT significantly improves the efficiency of picket administration, automates personnel strength recapitulation, and provides a real-time monitoring dashboard for leadership, while minimizing human error and ensuring data transparency and accountability.

Classification of Oil Palm Fruit Ripeness Levels Based on Digital Image Feature Extraction Using the Catboost Algorithm

2026-05-13T07:24:03+00:00

Determining the ripeness level of oil palm fruit is essential for improving palm oil production quality. Manual assessment methods are often subjective and inconsistent because they rely on workers’ experience and environmental conditions. Therefore, this study proposes an automatic image-based classification system using the CatBoost algorithm. The novelty of this research lies in the integration of CatBoost with RGB color and Gray Level Co-occurrence Matrix (GLCM) texture feature extraction for multiclass oil palm fruit ripeness classification. The dataset consisted of 1000 images categorized into four classes: unripe, under-ripe, ripe, and overripe. The research stages included image preprocessing, feature extraction, classification, and web-based implementation using the Flask framework. Experimental results showed that the proposed system achieved high performance based on accuracy, precision, recall, and F1-score metrics, demonstrating the effectiveness of CatBoost in classifying oil palm fruit ripeness while reducing overfitting. The developed web-based system can assist plantation workers in determining fruit ripeness automatically, objectively, and efficiently, thereb

Design and Development of an IoT-Based Ground Vibration Monitoring System with Landslide Potential Classification Using SVMDesign and Development of an IoT-Based Ground Vibration Monitoring System with Landslide Potential Classification Using SVM

2026-05-07T08:13:58+00:00

Landslides are natural disasters that can cause casualties, environmental damage, and infrastructure losses. One challenge in landslide mitigation is the limited availability of monitoring systems that can provide fast, real-time, and interpretable information about ground conditions. This study aims to design and develop an Internet of Things (IoT)-based ground vibration monitoring system with landslide potential classification using Support Vector Machine (SVM). The system uses an MPU6050 sensor to acquire ground vibration data and an ESP8266 microcontroller to transmit data through a Wi-Fi network to a server. The collected data are processed through preprocessing, windowing, feature extraction, and standardization before being classified into three condition categories: Safe, Alert, and Danger. The classification results are displayed on an LCD, visualized through a web dashboard, and used to activate a relay as a warning mechanism when a dangerous condition is detected. The testing results show reliable data transmission with a 100% delivery success rate in the observed test and an average latency of approximately 0.3 seconds. The SVM model achieved 99.79% accuracy, with high precision, recall, and F1-score for all classes. Therefore, the proposed prototype can support ground vibration monitoring and early landslide warning efforts more effectively.

The Role of E-Learning in Enhancing the Effectiveness of Learning in the Digital Age of the 21st Century

2026-06-29T15:23:33+00:00

The advancement of information technology has driven the transformation of education through the implementation of e-learning as a digital learning medium. This study aims to analyze the role of e-learning in improving learning effectiveness in the digital era of the 21st century. The research employed a library research method by reviewing secondary data from books, scientific articles, and national and international journals relevant to the topic. The data were analyzed using a qualitative descriptive approach through identification, classification, interpretation, and synthesis of findings from previous studies. The results indicate that e-learning contributes significantly to improving learning effectiveness by providing flexible access to learning resources, enhancing interaction, and fostering 21st-century competencies, including critical thinking, collaboration, communication, creativity, and digital literacy. However, its implementation still faces challenges, such as limited technological infrastructure, unequal internet access, and inadequate digital competencies. Therefore, supportive educational policies, improved technological infrastructure, and enhanced digital competencies of educators and learners are essential to ensure effective, high-quality, and sustainable learning in the digital era.

Real-Time Web-Based Indonesian Sign Language (BISINDO) Translator System Using CNN-LSTM Deep Learning and Text-to-Speech

2026-05-07T02:13:30+00:00

Indonesian Sign Language (BISINDO) is the primary communication medium for the deaf community, yet low public understanding often causes communication barriers. Previous sign language recognition studies mostly operated offline, lacked real-time web integration, and only produced text output. This study designs and develops a real-time web-based BISINDO translator system using a hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) method, integrated with a Text-to-Speech (TTS) feature. The dataset consisted of primary video data from 3 subjects, covering 11 category classes with 1,000 frames per class in grayscale format (100x89 pixels). The hybrid CNN-LSTM model was integrated into a React.js and Node.js web application (NzSignify). Testing results demonstrate that the model achieved 96% static accuracy based on Confusion Matrix evaluation. In real-time functional testing, an 80% Confidence Threshold effectively filtered incorrect gestures, enabling accurate translation of valid sign gestures into text and voice output.

Analysis Of The Distribution Of Livestock Disease Cases By Region Based On Data From The Ministry Of Trade’s Animal Health Center Using The Dbscan Clustering Method In Bandar District

2026-05-13T07:36:52+00:00

Livestock farming serves as a vital economic pillar for the community in Bandar District, Simalungun Regency. However, the high intensity of livestock activities is accompanied by a significant risk of disease transmission, which has historically been managed through conventional recording methods that lack spatial integration. This research aims to analyze the spatial distribution patterns of livestock diseases by implementing the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method integrated into a web-based Geographic Information System (GIS). Using a quantitative approach, the study processed 200 case records from December 2025 to January 2026. Spatial distances were calculated using the Haversine formula to ensure geographic accuracy. The results indicate that the optimal parameters for the DBSCAN algorithm are an epsilon ($\epsilon$) of 3.0 km and a minimum points (MinPts) of 2. These parameters successfully identified two primary clusters with zero noise, encompassing all 200 cases. Cluster 1 (98 cases) is concentrated in the west-central region, dominated by cattle and goats with diverse pathologies such as Scabies and BEF. Cluster 2 (102 cases) is located in the east-northern region and exhibits a more heterogeneous livestock profile, including rabies cases in dogs. High-density areas requiring priority intervention were identified in Pematang Kerasaan Rejo and Perdagangan II. The developed web-based GIS provides an interactive visualization platform that enhances early warning capabilities and supports data-driven decision-making for livestock disease surveillance and regional control.

Application System for Tracking Training and Physical Condition of Student Cadets in the Student Cadet Corps Regiment

2026-05-07T03:54:58+00:00

Monitoring physical condition and training activities is an important aspect in the development of student cadets within the Student Cadet Corps Regiment. However, the monitoring process that is still conducted manually reduces the effectiveness of tracking training compliance and the physical development of cadets. This study aims to design and develop a web-based application system for tracking the training activities and physical conditions of student cadets to support structured recording, monitoring, and evaluation of physical fitness. The system development method used in this research is Rapid Application Development (RAD), which emphasizes rapid prototyping and iterative improvements based on user feedback. The system was developed using the CodeIgniter framework for backend development, Bootstrap for frontend design, and MySQL as the database management system. The application enables cadets to record running activities and physical condition data such as height and weight, which are then calculated using the Body Mass Index (BMI) method. The data are stored in the database and visualized through graphical progress reports to facilitate monitoring of training outcomes. In addition, the system provides an administrative dashboard that allows regiment officials to centrally monitor the physical condition and training compliance of cadets. The results show that the developed system can improve the effectiveness and efficiency of training monitoring compared to the previously used manual method.