Implementation of Deep Learning in Science Education : A Bibliometric Analysis Based on Scopus Irdes Hidayana Siregar (a*), Festiyed (b), Minda Azhar (c), Asrizal (d)
a) Universitas Negeri Padang, Jalan Prof. Dr. Hamka Air Tawar Bar., Kec. Padang Utara, Kota Padang, Sumatera Barat 25171, Indonesia*irdeshidayana.sirega[at]gmail.com
b) Universitas Negeri Padang, Jalan Prof. Dr. Hamka Air Tawar Bar., Kec. Padang Utara, Kota Padang, Sumatera Barat 25171, Indonesia
Abstract
This study aims to analyze trends and developments in research related to the use of deep learning in science education through a bibliometric approach based on data from Scopus. The method used is bibliometric analysis with the assistance of VOSviewer software and Scopus analysis features to identify the number of publications, dominant keywords, document types, author collaborations, funding sponsors, and publication sources. The analysis results show that publications related to this topic have increased significantly from 2015 to 2025, with a dominance of scientific articles and conference proceedings. Keywords such as deep learning, science education, machine learning, and students occupy central positions in the co-occurrence network, indicating a strong connection between artificial intelligence technology and learning practices in the STEM field. Additionally, prominent funding agencies such as the National Science Foundation and the National Natural Science Foundation of China actively support research in this field. Although the direct application of deep learning to elementary school students remains limited, this technology holds great potential as a pedagogical tool for more adaptive, personalized, and data-driven science education. These findings open opportunities for the development of innovative learning media that can enhance the quality of science education, particularly in supporting active learning approaches and strengthening critical thinking skills. This research provides an important foundation for further research and the development of technology-based educational policies.
Keywords: Bibliometric analysis- Deep learning- Science Education- Scopus- VOSviewer
