Advancing biodiversity education: The role of deep learning in fish identification

Abstract

This study examines trends in deep learning applications for fish identification, highlighting its significance for biodiversity education and conservation. Accurate identification of fish species enhances students’ understanding of aquatic ecosystems and promotes environmental sustainability. A bibliometric approach was employed to analyze 737 articles published between 2014 and 2024, using keyword mapping, citation patterns, and inter-article linkages in the VOSviewer application. The analysis revealed that deep learning techniques, particularly Convolutional Neural Networks, improve the accuracy of fish species identification compared to traditional methods. Despite these advances, the integration of such technologies in educational contexts remains limited, representing a notable research gap. Based on these findings, the study advocates for the incorporation of deep learning tools into biodiversity education curricula to foster interactive, efficient, and technology-driven learning experiences. The research underscores the potential of leveraging advanced computational models to enhance both pedagogical practices and conservation efforts. By bridging artificial intelligence with environmental education, this study provides a framework for developing innovative strategies that support student learning, improve ecological literacy, and contribute to the sustainable management of fisheries resources.

Keywords: Biodiversity education; convolutional neural networks; deep learning; fish identification; technology integration