Deep learning and genomic strategies for ocean literacy development

Abstract

Indonesia, as a nation with rich marine biodiversity, faces significant challenges in fostering ocean literacy due to the limited integration of ecological and technological content in formal education. In this study, we aimed to evaluate ocean literacy among university students through a project-based learning program incorporating deep learning and genomic data for fish species identification. The initiative responded to the urgent need for educational reform that supports sustainable ocean conservation practices. Using a quantitative approach, the program engaged 36 undergraduate biology students in various activities, including marine problem analysis, field-based fish species identification via the Fishiden platform, phylogenetic tree construction using genomic data, and public outreach through social media advocacy. Pretest and post-test evaluations measured changes in ocean literacy, while Likert-scale questionnaires assessed student engagement and perception. The results showed a significant increase in post-test scores, with a mean improvement from 58.11 to 71.88, and a moderate N-Gain of 0.30. Students who demonstrated positive responses to the program also achieved higher gains in ocean literacy, indicating a strong correlation between engagement and learning outcomes. The highest evaluation scores reflected students' enhanced awareness, inquiry skills, and attitudes toward biodiversity conservation. Technological tools such as deep learning and genomic databases effectively support interactive and data-driven learning. These findings highlight the transformative potential of integrating technology into ocean literacy education. This research bridges technology, pedagogy, and marine conservation, emphasizing the integration of innovative educational strategies for sustainability. In the future, researchers should address study limitations and improve the accuracy of deep learning models.

Keywords

• conservation education
• deep learning
• genomic data
• ocean literacy
• species identification