| Popis: |
This research discusses the trends in machine learning (ML) applications for predicting ecosystem responses to environmental changes. A keyword search was conducted in the WoS database using Boolean operators to identify relevant peer-reviewed articles. The search focused on English-language documents published between 2014 and 2023, while excluding non-original articles. Bibliometric data, includingpublication trends, citation counts, author collaboration patterns, and keyword analysis, were extracted from 554 retrieved articles. The data was then analyzed and visualized using R and VOSViewer. The study highlights the significant growth in annual scientific production, reflecting a growing interest in thisinterdisciplinary field. Core concepts such as “climate change,” “biodiversity,” and “ecological responses” continue to receive significant attention, while contemporary themes like “variability,” “time-seriesanalysis,” and “organic matter” are emerging. Co-authorship networks demonstrate extensive collaborationsacross countries, with the United States and China playing prominent roles. The research topics have evolvedfrom “ecological responses” and “community” to a focus on “model,” “optimization,” and “performance,” with an emphasis on fine-tuning models to incorporate climate variability. |
