Quantifying the research convergence of optimized generative adversarial networks with genetic algorithm using scientometric review analysis

Genetic Algorithms (GA) and Generative Adversarial Networks (GANs) have become increasingly popular in recent years due to their applicability in the field of generative modeling, adaptive optimization, and adversarial learning. This paper provides a scientometric review of the literature in GA–GAN convergence by analyzing its growth in publication, intellectual structure, patterns of collaboration, and thematic development. The final corpus comprised 276 peer-reviewed publications that were published between 2017 and 2025 and retrieved using bibliographic records that were found in Scopus and Web of Science. To assess the performance indicators and science-mapping patterns, the Bibliometrix/Biblioshiny and VOSviewer were used to analyze the data. The findings reveal that the GA–GAN studies grew significantly since 2020, with the maximum indexed yearly output recorded in 2024, and the 2025 number simply indicates a partial-year record at the moment of data gathering. It is structured around a methodological core, consisting of evolutionary optimization, adversarial generative modeling, and general ideas of deep-learning, with uses spanning to forecasting, security, image analysis, biomedical settings, urban design, and structural modeling. The results also suggest that the activity of publication and its impact is localized in a few countries, institutions, authors, and repetitive sources, whereas international collaboration is relatively small. In general, the review indicates that GA–GAN studies hold unambiguous methodological and cross-domain potential, yet its development is limited due to repetitive issues such as computational cost, instability in generative optimization, evaluation inconsistency, and lack of empirical maturity across domains. The study provides a systematic literature review of the discipline and determines future directions regarding the scalability, robustness, interpretability, and wider domain-specific validation.