March 2025
·
79 Reads
International journal of Environmental Science and Technology
This study explores the application of machine learning techniques to predict plant species richness (PSR) on islands, addressing the challenges of traditional modeling approaches. Eighteen machine learning models were compared using the PyCaret library, with Gradient Boosting Regressor (GBR) emerging as the most accurate predictor for PSR across 20 Nile River islands. The GBR model achieved impressive results, with metrics like R 2 of 0.996 and MAE of 0.376, demonstrating its effectiveness in capturing the complex relationship between environmental factors and plant diversity. The research highlights the significant influence of climate factors, particularly minimum temperature on PSR. They account for 60.3% of the variation in species richness. This comprehensive study showcases the potential of machine learning to significantly enhance our ability to predict and understand plant diversity patterns and it is considered one of the most comprehensive examinations of plant diversity trends to date.