Claim Missing Document
Check
Articles

Found 38 Documents
Search

Application of Ant Colony Algorithm to Determine the Shortest Route for Nature and Culinary Tourism in North Aceh Muhammad Teguh; Wahyu Fuadi; Zahratul Fitri
International Journal of Engineering, Science and Information Technology Vol 5, No 2 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i2.857

Abstract

This Research aims to design and implement a shortest route determination system for natural and culinary tourism locations in North Aceh using the Ant Colony Optimization (AntCO) algorithm. The developed system is designed to help tourists plan their trips efficiently by considering the distance and travel time between tourist destinations. The system implementation using the AntCO algorithm successfully displayed optimal routes for 28 tourist destinations in North Aceh. The system successfully implemented filtering features based on tourism categories and route visualization on an interactive map using different markers (green for natural tourism and red for culinary tourism). The research results show that the system successfully optimized tourist travel routes and provided comprehensive information, including automatic location detection, a list of tourist destinations, travel route details, and optimal visit sequences based on selected tourism categories. This system proved effective in helping tourists plan their trips in North Aceh by providing efficient routes according to their preferred tourism category preferences.
Design of Attendance System for Informatics Engineering Lecturers Using RFID Sensors Based on IoT and Telegram Applications Andry Maulana Akbar; Wahyu Fuadi; Nunsina Nunsina
International Journal of Engineering, Science and Information Technology Vol 5, No 2 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i2.794

Abstract

The attendance system is an essential element in the academic environment to ensure lecturer attendance in the lecture process. However, the manual attendance method still has various weaknesses, such as the potential for data manipulation and inefficiency in recording attendance. To overcome these problems, this research designs and implements an Internet of Things (IoT)-based lecturer attendance system using Radio Frequency Identification (RFID) sensors integrated with the Telegram application. The research method includes hardware design with ESP32 microcontroller, ESP32-CAM, RFID sensor, and HC-SR04 ultrasonic sensor. This system works by detecting lecturer attendance through RFID cards confirmed by ESP32, taking pictures with ESP32-CAM, and sending automatic notifications via the Telegram bot. Lecturer attendance data is then stored in a web-based database to facilitate the monitoring and evaluation. The test results show that the developed system can detect and record lecturer attendance accurately, with the response speed of the RFID sensor in reading cards ranging from 1-5 cm. The ultrasonic sensor also successfully detects objects accurately within a predetermined distance range. Lecturer attendance notifications sent via Telegram allow administrators to conduct real-time monitoring. With this IoT-based attendance system, the attendance recording process becomes more efficient and transparent and can reduce the risk of data manipulation. Further development can be done by adding data encryption and biometric authentication features to improve system security.
Clustering Agricultural Productivity by Type and Results Using K-Medoids Method in Districts North Aceh Mutia Zahara; Wahyu Fuadi; Rini Meiyanti
International Journal of Engineering, Science and Information Technology Vol 5, No 1 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i1.699

Abstract

This research aims to develop a web-based application that can cluster sub-districts in North Aceh District based on the type and yield of agricultural productivity, focusing on increasing the ease of visualization and data analysis by users. The method applied in this research is K-Medoids, a clustering technique used to group sub-districts based on high, medium, and low harvest levels. The application will use data from the North Aceh District Agriculture Office, covering 2021 to 2023, including various food crops such as rice, corn, peanuts, green beans, cassava, sweet potatoes, and soybeans. This research will analyze the sub-district name, type of agriculture, year of production, planting area, and harvest area to identify clusters of sub-districts with similar agricultural yield patterns. The system is developed using the PHP programming language to facilitate implementation and data access by stakeholders. As an evaluation tool for clustering results, the Davies-Bouldin Index (DBI) is used to measure the quality of clustering results. The results of this study are expected to provide insights into agricultural productivity in North Aceh District and assist policymakers in designing more effective strategies to increase agricultural yields, especially in low-yielding sub-districts. In addition, this application also provides an interactive platform for users to analyze agrarian data quickly and efficiently.
Cataract Eye Disease Diagnosis Using the Random Forest Method Lilis Novita; Wahyu Fuadi; Kurniawati Kurniawati
International Journal of Engineering, Science and Information Technology Vol 5, No 2 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i2.777

Abstract

This study developed a machine learning-based classification model using the Random Forest algorithm to detect cataract risk based on 11 variables: age, gender, family history, lens opacity, visual acuity reduction, light sensitivity, color changes, double vision, intraocular pressure, slit-lamp results, and visual acuity. Feature importance analysis revealed that lens opacity and visual acuity variables contributed most significantly to cataract risk prediction, followed by intraocular pressure and visual acuity reduction. The system was designed using Google Colab for model training and Streamlit as an interactive interface, enabling real-time predictions with intuitive result visualization. After optimization using Grid Search, the model achieved an accuracy of 92.0%, precision of 95.0%, sensitivity of 90.0%, F1 Score of 92.4%, and specificity of 98.0%. This system is expected to serve as an effective supporting tool for medical professionals in the early diagnosis of cataracts.
Application of Off-Grid Solar Panels System for Household Electricity Consumptions in Facing Electric Energy Crisis Selamat Meliala; Saifuddin Muhammad Jalil; Wahyu Fuadi; Asran Asran
International Journal of Engineering, Science and Information Technology Vol 2, No 1 (2022)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1122.382 KB) | DOI: 10.52088/ijesty.v2i1.199

Abstract

At this time the cost of electricity is very expensive which is felt by the community because the government is still exploring oil and natural gas which is the need for non-renewable energy sources that are running low. This non-renewable energy still dominates for power generation in the thousands of Mega Watts. To anticipate the problem of non-renewable energy that is so big, you can use the On Grid-Tie System, sunlight is converted into DC voltage through the Solar Module, a pure DC voltage that comes out of the solar module. And Off Grid-Tie System namely sunlight is converted into DC voltage through the Solar Module, pure DC voltage generated from the solar module. Then the pure DC voltage uses a DC to DC regulation module or is called a DC regulator. DC regulator which aims to regulate the storage of DC current into the battery. Then the battery is used to supply power to the inverter. The method used in this study uses an off-grid solar home system as a power supply for households that are far from the electricity network or save electricity consumption due to expensive electricity rates. For settings for the intensity of sunlight using a portable holder, the solar panels are shifted manually in order to get the optimal light intensity to produce large output power. In off-grid application testing at household loads, from a load test of 93.5 watts to 750 watts, it shows that the load current is getting bigger and the discharging current is also large so that the duration of using the off-grid system from a load of 93.5 watts is 6 hours long and at a load of 750 watts. up to 15 minutes. This is because the condition of the lead-acid battery is maintained from 13.56 to 11.5 Volt DC, the battery should not be forced below the 11.5 Volt voltage because it will cause damage to the battery. For the use of loads that respond to very high instantaneous currents such as electric irons, dispensers, rice cookers should use more batteries and use an off-grid system voltage higher than 12 Volt DC.
Student Learning Style Decision-Making System Using the Multi-Attribute Utility Theory Method at SMA Negeri 1 Jangka Munawarah Munawarah; Wahyu Fuadi; Hafizh Al Kautsar Aidilof
International Journal of Engineering, Science and Information Technology Vol 5, No 2 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i2.842

Abstract

Education plays a vital role in shaping individual development and national progress. One key factor influencing learning effectiveness is students' learning styles, which determine how individuals absorb, organize, and process information. Understanding these differences is crucial for designing effective teaching methods. This research develops a Decision Support System (DSS) to determine student learning styles at SMA Negeri 1 Jangka using the Multi-Attribute Utility Theory (MAUT) method. MAUT is chosen for its ability to evaluate multiple criteria, convert them into numerical values, and systematically identify the most suitable learning approach. The alternatives in this study include Project Based Learning (PBL), Problem-Based Learning (PrBL), Inquiry-Based Learning (IBL), Discovery Learning (DL), and Contextual Teaching and Learning (CTL). The MAUT analysis considers five criteria: student activeness, material understanding, collaboration, initiative and creativity, and teacher-student communication. The research stages include literature study, data collection, system and database design, MAUT implementation, and system evaluation. The results, based on MAUT calculations, show that Inquiry-Based Learning (IBL) scores the highest at 13.611, followed by Discovery Learning (DL) at 13.018, Problem-Based Learning (PrBL) at 12.975, Contextual Teaching and Learning (CTL) at 12.929, and Project Based Learning (PBL) at 12.558. This system assists educators in designing personalized learning strategies that align with students' strengths. Leveraging data-driven analysis enhances education quality, fosters a student-centred learning environment, and improves academic performance and lifelong learning habits.
Comparative Analysis of XGBoost and CatBoost for Detecting AI-Generated Synthetic Faces Using Texture and Color Features indra firmansyah; Defry Hamdhana; Wahyu Fuadi
Journal of Deep Learning, Computer Vision, and Digital Image Processing Volume 4 Issue 3 September 2026
Publisher : CV. Sakura Digital Nusantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61255/decoding.v4i3.1677

Abstract

Purpose – The rapid advancement of Generative AI produces near-perfect synthetic face images, posing fraud risks. This study compares XGBoost and CatBoost for detecting real versus synthetic faces using texture and color features.Methods – Texture features were extracted using GLCM with four orientations (0°, 45°, 90°, 135°), generating 16 features, while YCbCr mean and standard deviation provided 6 color features (22 total). A balanced 750-image dataset (real and AI from GPT Image 2, Nano Banana, Leonardo AI, Canva, Dreamina AI) was split 80:20 and normalized with StandardScaler.Findings – On an asymmetric, source-shifted test split, where real test images were captured independently while synthetic test images came from the same generators as training, CatBoost outperformed XGBoost, achieving 84.67% accuracy (vs. 80.67%), 83.33% precision (vs. 78.05%), 86.67% recall (vs. 85.33%), and 84.97% F1-score (vs. 81.53%), reducing False Positives by 27.78% and False Negatives by 9.09%. However, repeated 5-fold stratified cross-validation (5×10 repeats) revealed only marginal, non-significant differences (0.05-0.87 percentage points, p > 0.09). Research implications – Although the moderate dataset excludes recent architectures such as StyleGAN, findings affirm the feasibility of efficient, interpretable detection through feature engineering. Since within-domain cross-validation shows no significant algorithmic difference, the observed cross-domain gap should be regarded as preliminary, pending validation through repeated cross-domain evaluations across diverse AI platforms.Originality – The novelty lies in comparing XGBoost and CatBoost on multi-orientation GLCM and YCbCr feature fusion for contemporary AI-generated images, identifying cr_std and contrast_0 as stable, interpretable discriminative markers.
Analysis of Hoax News in News Media Using the K-Nearest Neighbor (K -NN) Method Putri Nabila; Bustami; Wahyu Fuadi
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 6 No. 1 (2026): October 2026
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v6i1.2799

Abstract

The rapid development of internet technology has facilitated the dissemination of information within the community through various news media. However, this convenience has also fueled the widespread dissemination of unverified fake news (hoaxes), potentially causing social unrest, undermining the credibility of information, and manipulating public opinion. Therefore, a computing system capable of automatically filtering and categorizing news is needed to help readers distinguish between facts and hoaxes. This study aims to develop an automatic classification model to identify hoax news using the K-Nearest Neighbor (KNN) method and measure its accuracy. This study uses the Euclidean Distance approach with a value of k = 3. This study uses 2,000 news data that have been collected and verified through official fact-checking sites. The data was divided using the train-test split method with a ratio of 80:20 to obtain 1,600 training data and 400 testing data . Based on the results of the confusion matrix test at a value of K = 3, the True Positive (TP) value was 104, False Positive (FP) was 0, False Negative (FN) was 95, and True Negative (TN) was 201. The model performance evaluation showed a precision value of 100%, recall of 52.26%, and accuracy of 76.25%. These results indicate that the K-NN method is able to classify news well and has a high level of precision, although the model's ability to detect all hoax news ( recall ) still needs to be improved in further research. This research is expected to contribute as a technology-based tool ( decision support system ) for the public and academics in monitoring, filtering, and tackling the spread of fake news in digital media.