Intelligent assessment of habitat quality based on multiple machine learning fusion methods

Kui Yang, Dongge Cui, Chengrui Wang, Qi Tang, Linguang Miao,
Intelligent assessment of habitat quality based on multiple machine learning fusion methods,
Engineering Applications of Artificial Intelligence,
Volume 162, Part A,
2025,
112395,
ISSN 0952-1976,
https://doi.org/10.1016/j.engappai.2025.112395.
(https://www.sciencedirect.com/science/article/pii/S0952197625024030)
Abstract: Evaluating habitat quality can help balance the relationship between economic development and biodiversity conservation, and it serves as a foundation for constructing an ecological security pattern. However, research on the intelligent construction of habitat quality is limited. This study develops a comprehensive framework to assess habitat quality based on optimized machine learning methods. The findings of the research are as follows: (1) From the perspective of human-machine interactive interpretation, ensemble learning is used to enhance the performance of basic classifiers, resulting in a classification map with high precision and recall. (2) The particle swarm optimization (PSO) algorithm can improve the goodness of fit of the Extreme Gradient Boosting (XGBoost) inversion model by 4–5 %. (3) The habitat quality inversion method based on XGBoost-PSO has high credibility and application value, with its texture structure being the result of both expert experience and image information interaction. (4) The model demonstrates certain application potential in downscaling; under the seven-band perspective, the blue and near-infrared bands are the most important, while in the four-band perspective, green and near-infrared bands take precedence.
Keywords: Habitat quality; Ensemble learning; Extreme gradient boosting (XGBoost); Particle swarm optimization; Downscaling