<![CDATA[Deep learning approach can be used for personalized, adaptive education systems. Personalized learning can be utilized to make this type of education more effective. Personalized learning (also known as competency-based learning) is an educational approach that adapts instruction to students' needs and abilities. This research aims to identify and analyze research trends of deep learning in physics. This research method is descriptive and analytical. The data used in this research was obtained from documents indexed by Google Scholar from 2016-2025 using Publish or Perish and Dimension.ai. Research procedures use PRISMA guidelines. The data identified and analyzed are the type of publication, publication source, and the title of research on deep learning in physics that is widely cited. The data analysis method uses bibliometric analysis assisted by VOS viewer software. The results of the analysis show the research trend on the deep learning in physics indexed by Google Scholar from 2016 to 2025 has experienced a fluctuating increase. However, in 2025 there will be a decline in the research trend on the deep learning in physics. There are many documents in the form of articles, proceedings, chapters, preprints, monograph and edited books that discuss research into the deep learning in physics. Key words that are often used in research about it are chatgpt, technology, learner, artificial intelligence, etc. The study results show that the implementation of deep learning has a positive impact on the development of higher-order thinking skills, such as critical thinking, creativity, and problem-solving, which are highly relevant to the demands of 21st-century competencies. However, its effectiveness depends heavily on the quality of the learning design, teacher preparedness, and an assessment system that supports meaningful learning.]]>
Copyrights © 2026
