Impact of Mathematical Optimization on Machine Learning: A Bibliometric Review

The article "Influence of mathematical optimization in machine learning: a bibliometric review" analyzes the contributions, scientific trends, and influence of mathematical optimization in machine learning. Through a bibliometric analysis, 67 articles selected from Scopus between 2013 and 2024 were reviewed. The results show a rapid growth in publications from 2019, reaching a peak in 2023, from which it can be inferred that it is important to articulate and integrate mathematical optimization techniques into algorithms for machine learning. We highlight the publications of the journals IEEE Access and Computers and Operations Research, as well as international collaboration (97% of co-authorships), with key institutions such as the University of Seville and funding from the National Science Foundation. Driving topics such as deep learning and integer programming have led semantic development, while emerging areas include data mining and optimization in human resources. Bradford's Law and semantic analysis corroborate the thematic concentration in journals and the evolution towards advanced techniques. Additionally, the bibliometric review evidences many practical applications, including resource management and energy optimization. The limitations of the analysis focus on the exclusive use of Scopus; it is suggested to expand the study to other databases to capture a broader picture and delve deeper into specific applications. This work consolidates mathematical optimization as an essential axis in the advancement of machine learning, opening up opportunities for interdisciplinary innovations.