RANCANG BANGUN SISTEM PEMANTAUAN HOLISTIK BERBASIS INTEGRASI DATA MULTIMODAL (STUDI KASUS : SD AL-BIRU)

Problems in monitoring students at the elementary school level, especially in academic, psychosocial, and attendance aspects, are still dominated by a manual approach that is not integrated, resulting in late intervention, potential data falsification, and inefficiency in verifying grades and diplomas. This study aims to develop and implement an artificial intelligence-based mobile and web application system called the Student Wellness Tracker, which is designed as a holistic student monitoring platform.
This system utilizes multimodal input in the form of academic grade data, facial expressions, partcipacing, and attendance, to detect student risks early through predictive analysis. The system development method uses the Vue.js and Tailwind CSS frameworks on the frontend side, and PHP and Firebase for backend management and data authentication. The AI component is developed with TensorFlow.js to detect facial expressions in real-time on the client side (browser) and CNN algorithm for detect the face emotion. The expected results of this system are increased teacher efficiency in identifying and handling student problems, providing specific intervention recommendations based on data.
The conclusion of this study shows that the integration of AI technology in student monitoring systems is able to produce adaptive solutions that support faster, more precise, and personalized decision making in the context of elementary education.
Keywords: student monitoring, artificial intelligence, multimodal analysis, facial expression detection, risk prediction, mobile applications, elementary education.