Development of AI-Based Motion Detection System Media for Kinematics Experiments to Improve Students' Understanding of Straight Motion Concepts and Data Analysis Skills.

Abstract

The development of artificial intelligence (AI) technology provides significant opportunities for improving the quality of science learning, particularly physics. This study aims to develop a kinematics experiment media based on an AI motion detection system to improve students' understanding of linear motion concepts and data analysis skills. The research method uses a bibliometric approach by analyzing publications in the Scopus database for the period 2006–2025. Data were collected through article metadata extraction including title, abstract, keywords, author names, publication year, and journal source, then analyzed using the Bibliometrix package in R software and the Biblioshiny interface. The analysis includes annual scientific production trends, the most relevant publication sources, contributions per country, and thematic maps. The research results show a significant increase in publications related to the integration of AI in physics learning, dominated by countries with high research capacity such as China, India, and the United States. Research themes are still dominated by technical aspects such as computer vision, object detection, and video surveillance, while direct application to physics learning contexts, particularly kinematic experiments, is still limited. These findings highlight a research gap in the need to develop AI-based learning media that not only improve measurement accuracy but also support students' data analysis skills. This research contributes theoretically by enriching the literature on the use of AI in science education, and practically by offering alternative innovative learning media for more effective and interactive linear motion experiments.