JICDA: Journal of Informatics, Computer Science, Data Science And Artificial Intelligence

IMPLEMENTATION OF GAUSSIAN NAÏVE BAYES ALGORITHM FOR CLASSIFICATION OF BUMDESMA FINANCIAL HEALTH LEVELS

Annisa Rahma Lubis, Darjat Saripurna, Tasliyah Haramaini — Sun, 07 Dec 2025 00:00:00 +0700

Joint Village Owned Enterprise (BUMDesma) is a business entity established by two or more villages to strengthen local economic cooperation. In Batu Bara Regency, BUMDesma Cipta Karya Srikandi in Talawi District and Datuk Tanah Datar District has shown significant development with assets reaching Rp3.6 billion and more than 143 active member groups. However, BUMDesma's financial management still faces obstacles such as unsystematic recording and non-objective evaluation of financial reports. This research aims to implement the Gaussian Naïve Bayes algorithm to classify the financial health level of BUMDesma based on financial ratios from SPP loan development reports during the 2022-2024 period. This algorithm was chosen due to its ability to handle numerical data and produce efficient classification models. The research results show that from 60 sample data, 50 data were obtained in the healthy class with a prior probability of 83.33%, mean value (% return) of 94.72, and standard deviation of 2.87, while 10 data in the unhealthy class had a prior probability of 16.67%, mean value (% return) of 85, and standard deviation of 3.46. These classification results can serve as a reference for BUMDesma managers in decision-making related to fund management.

IMPLEMENTATION OF GEOFENCING ALGORITHM FOR MOBILE-BASED ONLINE STUDENT ATTENDANCE SYSTEM AT SMA NEGERI 5 MEDAN

Dio Kiasta, Mhd Zulfansyuri Siambaton, Tasliyah Haramaini — Sun, 07 Dec 2025 00:00:00 +0700

The manual attendance system still used in many schools has significant weaknesses such as time inefficiency, vulnerability to "proxy attendance" manipulation, and human error in the recapitulation process. This research aims to develop a mobile-based student attendance system with Geofencing technology using the AABB (Axis-Aligned Bounding Box) algorithm to ensure the validity of student attendance only within predetermined school areas. The research method employs a Prototyping approach with a technology stack consisting of Flutter for the frontend, Laravel for the backend API, MySQL and Firebase Realtime Database for data storage, and Mapbox API for 3D map visualization. The novelty of this research lies in the implementation of a 3D volumetric virtual fence that validates not only horizontal coordinates (latitude and longitude) but also the vertical altitude of students, addressing manipulation vulnerabilities present in conventional 2D Geofencing systems. Functional testing results (black box testing) demonstrate that all main system features operate according to specifications for the three user roles (Administrator, Teacher, and Student). Geofencing accuracy testing proves that the AABB algorithm is capable of validating locations with a 100% success rate across various test scenarios, including rejection of attendance attempts from outside the configured horizontal and vertical boundaries. Compatibility testing on 20 Android devices shows that the application runs stably with good location accuracy. This system successfully enhances attendance process efficiency, attendance data accuracy, and significantly prevents location manipulation. The implementation of Firebase Realtime Database enables real-time attendance monitoring by teachers, while the Mapbox 3D-based visual interface facilitates Administrator configuration of validation areas. This research provides an important contribution to the development of Location-Based Services (LBS) systems for the education sector through a three-dimensional spatial validation approach.

IMPLEMENTATION OF AUGMENTED REALITY (AR) TECHNOLOGY FOR PRODUCT CATALOG AT GABE CRAFT MSME

Khoirunnazwa Panggabean, Khairuddin Nasution, Rachmat Aulia — Sun, 07 Dec 2025 00:00:00 +0700

UMKM Gabe Craft, as a handicraft business, requires a more interactive promotional medium to increase marketing reach and product appeal. This research aims to design and develop an Augmented Reality (AR) based product catalog application as a modern, informative, and attractive digital promotional solution. The system is developed as a website using WebAR technology that allows users to display 3D product models through QR Code scanning without needing to install additional applications. The development process uses a web based development approach that includes interface design, 3D model integration, AR feature implementation, and system testing. AR technology is implemented using Model Viewer to display 3D product models in real-time, enabling users to view product details such as shape, size, and appearance more interactively. Testing results using the black box method show that all system functions work properly according to requirements, including catalog display, 3D model visualization, and product data management by admin. Based on the research findings, the implementation of Augmented Reality in UMKM Gabe Craft's product catalog has proven capable of improving promotional quality, providing new experiences for consumers, and supporting the digitalization process of MSMEs

AUGMENTED REALITY TECHNOLOGY IMPLEMENTATION FOR JEPARA CARVING PRODUCTS AS A PROMOTIONAL MEDIUM AT SAMUDRA PUTRA FURNITURE STORE

M. Rofi Fahrozi, Khairuddin Nasution, Tasliyah Haramaini — Sun, 07 Dec 2025 00:00:00 +0700

The furniture business competition in the digital era faces challenges in promotion, namely the inability of 2D media to present product scale and proportion visualization accurately. This limitation creates visual uncertainty (customer hesitation) that hinders purchasing decisions. This research aims to design and develop an Augmented Reality (AR)-based interactive promotional medium to provide a virtual try-before-you-buy experience. The development methodology employs the Multimedia Development Life Cycle (MDLC) with a backend interface for product management (CRUD) and a frontend for product catalog and AR visualization. The implemented technologies include AR as the core of 3D visualization, Quick Response Code (QR Code) as an access trigger, and WebAR as a browser-based platform. System testing utilizes Black Box Testing methods covering functional testing, AR quality, and User Acceptance Testing (UAT). The research results demonstrate that the system was successfully built with 100% functionality operating according to specifications. AR functionality proved capable of projecting 3D models stably and interactively in the user's environment. UAT results show that the system is well-accepted by the store for managing products and by prospective customers for assisting the visualization process. The novelty of this research lies in the implementation of markerless WebAR with plane detection that enables direct access through browsers without application installation, as well as a focus on digitizing Jepara carving products with high visual complexity. This research concludes that AR implementation is a functional and valid solution to overcome 2D media visualization problems and has the potential to increase customer confidence in making online furniture purchasing decisions.

IMPLEMENTATION OF FISHER-YATES ALGORITHM IN MOBILE-BASED ONLINE EXAM APPLICATION AT SMP NEGERI 1 KARIMUN

Nur Alya Afifah, Mhd Zulfansyuri Siambaton, Rachmat Aulia — Sun, 07 Dec 2025 00:00:00 +0700

The development of information technology has driven significant changes in educational evaluation systems, one of which is through the implementation of mobile-based examinations. However, one of the main challenges in conducting online examinations is maintaining question randomization so that distribution is fair and reduces the potential for cheating. The Fisher-Yates algorithm is an algorithm used to shuffle the order of elements in a list or array in a perfectly random (uniform random) manner, where every possible permutation of those elements has an equal probability of occurring.

The implementation results demonstrate that the Fisher-Yates algorithm is capable of producing evenly distributed and non-repetitive question randomization, thereby enhancing the fairness and reliability of the examination system. Thus, the application of this algorithm can support the effectiveness of the learning evaluation process and provide a positive contribution to the utilization of educational technology at the junior high school level.