The Use of Deep Learning Algorithms in Industry: A Bibliometric Evaluation

With the acceleration of the digital transformation process, the effective analysis of large volumes of complex data generated by industrial systems has become a critical necessity. In this context, deep learning algorithms have begun to be widely used within the scope of Industry 4.0, thanks to their multi-layered structures and high representational capabilities. This study aims to examine the areas of application and development trends of deep learning algorithms in industrial applications using bibliometric analysis methods. Academic publications from 2000 to 2024 containing the keywords ‘deep learning’ and ‘industry’ in the Scopus database were analysed within the scope of the research. The historical development of the literature, publication trends, and prominent concepts were visualised using VOSviewer software; the studies were classified under thematic headings such as production automation, quality control, predictive maintenance, and autonomous systems. The findings reveal a marked increase in publications on deep learning, particularly after 2015, and highlight the prominence of models such as convolutional neural networks (CNN), LSTM, and GRU in industrial applications. Consequently, this study demonstrates that deep learning strengthens industrial decision-making processes, increases production efficiency, and offers the potential to reduce costs; it provides a literature-based framework for future applied studies.