The growing concern over marine biodiversity loss has accelerated the need for efficient and accurate aquatic species identification systems. Traditional monitoring methods are often time-consuming, expensive, and reliant on manual expertise. This paper presents M.A.R.L.IN: eDNA Species Classifier (Marine Analytics for Research, LIfe & IN sights), a machine-learning-based system designed to identify marine and aquatic species from environmental DNA (eDNA) data. The proposed model uses a deep learning-driven backend integrated with automated preprocessing pipelines to analyze genetic data collected from marine environments. It leverages model architectures trained to classify species accurately from DNA sequences while minimizing human intervention. The system’s modular design, powered by Python-based APIs, supports extensibility and integration with bioinformatics datasets. Experiments demonstrate that M.A.R.L.IN improves species recognition accuracy and processing efficiency over conventional statistical and rule-based methods. This approach paves the way for scalable, AI-assisted biodiversity assessment and conservation strategies in marine research. Furthermore, the system emphasizes real-time adaptability and cross-platform deployment for diverse research environments. Its architecture ensures efficient data flow from collection to inference, promoting high accuracy and reduced latency in prediction. M.A.R.L.IN also supports continual learning, enabling the model to evolve as new eDNA samples are introduced into the ecosystem. The integration of automation and explainable AI modules enhances result interpretability for biologists and environmental researchers. Overall, this project contributes to the advancement of intelligent marine monitoring systems, bridging the gap between artificial intelligence and ecological conservation.
Hemalatha et al. (Fri,) studied this question.