Тип публикации: статья из журнала
Год издания: 2025
Идентификатор DOI: 10.31407/ijees15.636
Ключевые слова: biodiversity, machine learning, environmental education, environmental risks, natural ecosystem
Аннотация: The contemporary biodiversity crisis requires qualitatively new approaches to the training of environmental specialists who would be proficient with tools for the digital analysis of data from protected natural areas. The purpose of the study was to develop and evaluate the effectiveness of an educational approach based on the use Показать полностьюof artificial intelligence tools to develop the competencies of environmental science students in the field of analyzing data from specially protected natural areas (SPNA). The pedagogical experiment involved 48 3rd-year students randomly assigned to the experimental (n = 24) and control (n = 24) groups. The experimental group underwent three-stage training to work with the Wildlife Insights platform to automatically identify animals in images from camera traps. The effectiveness of the approach was assessed by comparing the level of competency development, the quality of data analysis, and environmental thinking. The results showed a statistically significant superiority of the experimental group: the median total test score was 18.0 against 11.0 in the control group (U = 32.0, p < 0.001), the accuracy of species identification was 95.0% vs 78.0% (U = 89.5, p < 0.001), and task completion time was 2.9 times shorter. Students in the experimental group were more likely to identify complicated ecological patterns (87.5% vs 29.2%) and showed greater readiness to use digital tools in professional practice. The developed approach can be scaled for various environmental education programs, contributing to the goals of the Kunming-Montreal Framework for the effective management of protected natural areas.
Журнал: International Journal of Ecosystems and Ecology Science
Выпуск журнала: Т. 15, № 6
Номера страниц: 293-300
ISSN журнала: 22244980