Systematic Literature Review: Deep Learning Research in Physics Education at the Junior High School Level

Abstract

The advancement of artificial intelligence has increased research on deep learning in junior high school physics education, yet comprehensive reviews of its trends and implementation remain limited. This study aims to analyze publication trends, research focuses, implementation forms, and challenges of deep learning research in physics education over the past four years. The study employed a Systematic Literature Review (SLR) method involving the identification, selection, and synthesis of articles based on predefined inclusion and exclusion criteria. The findings indicate that deep learning research mainly focuses on improving conceptual understanding, reducing misconceptions, enhancing learning motivation, and developing data-driven adaptive learning systems. Its implementation has evolved from simulation-based applications and neural network–based diagnostic systems to the integration of augmented reality, educational chatbots, and learning analytics to support personalized and real-time learning. The study highlights that successful adoption depends on technological readiness, teacher pedagogical capacity, and digital infrastructure.