Akademik Veri Yönetim Sistemi
Aquaculture, cilt.612, 2026 (SCI-Expanded)
Machine learning (ML) is transforming aquaculture by supporting the development of knowledge-based, intelligent, and flexible production systems. As a subset of artificial intelligence, techniques such as deep learning and neural networks are increasingly integrated into data-driven aquaculture systems. This study presents a detailed overview of the progress in ML applications within aquaculture through a bibliometric analysis of “Articles” indexed in the Web of Science up to 2025, using the Bibliometrix package in R software. The analysis reveals a strong annual increase in publications (15.63 %), reflecting growing academic interest in the field. Thematic examination highlights a clear trend toward real-time monitoring, species detection, and predictive modeling, with frequently used terms such as “growth”, “system” and “mortality” forming the focus areas of current research. To explore possible future directions, six time series forecasting models (ARIMA, ETS, Holt-Winters, Polynomial Regression, Naive, and Theta) were applied. Among them, ARIMA estimates that the number of yearly publications could go beyond 990 by 2030. These findings offer useful insights into the ongoing expansion of the field and underline the important role of ML in promoting sustainable aquaculture practices.