<![CDATA[This study aims to evaluate the development, topic interconnections, and global research directions in the field of Deep Learning during the period 2019–2024, while also examining its implications for teaching statistical mathematics in the digital era. A bibliometric approach was used to analyze publication trends, citation patterns, and keyword relationships with the assistance of VOSviewer software. Data were obtained from the Scopus database using the main keywords “Deep Learning,” “Neural Networks,” and “Artificial Intelligence.” The results indicate that peak research activity occurred in 2022 with a significant surge in citations, followed by a decline in 2023–2024, marking a phase of research stabilization. Network analysis revealed that topics such as computer vision, medical imaging, and unsupervised learning dominate, while emerging trends like federated learning and edge computing are beginning to develop toward privacy and computational efficiency. Geographically, the United States and China are the main contributors to scientific publications, followed by Germany, the United Kingdom, and Australia. These findings highlight that the core success of Deep Learning is fundamentally grounded in statistical mathematics, particularly in optimization and probabilistic modeling. Accordingly, the implications for teaching statistical mathematics involve reorienting curricula toward applied, data-driven contexts emphasizing probabilistic thinking, algorithmic reasoning, and the integration of computational tools. Such an approach encourages students to bridge theoretical understanding with real-world problem solving in artificial intelligence and data science.]]>
Copyrights © 2026
