Utilizing data science to assess native Indian freshwater fish taxa and their conservation status

India ranks ninth globally in freshwater fish diversity but lacks updated checklists and data-driven fish diversity and conservation assessments, which are essential for better informed decision-making on conservation and species discovery efforts. This article proposes a scalable data science framework for data-driven assessment of freshwater fish taxa and their conservation status. The framework integrates automated checklist preparation form online resources, suitable analytics, and subsequent interpretation, which facilitates efficient policy planning on Native Indian Freshwater Fish (NIFF). The automated web harvesting component updates the checklist of 1239 NIFF species with minimal manual effort, which empower fish taxonomist to prepare accurate and periodic update. Newly introduced discovery analytics indicate a national temporal discovery rate of 18.86 per year for the period 1999-2024, which is the highest among four periods of discovery since 1758. And Northeast India contributed the highest biogeographic species discovery rate of 8.3 per year. The study identifies unknown population status of 76 % of NIFF, which hinder IUCN 'Red List' assessments. The analytics phase demonstrates the strength of the Machine Learning (ML) modelling for threat risk assessment of NIFF, achieving 84 % accuracy in predicting threat status. The Gradient Boosting Tree ML model identifies threat risk factors in order of 'habitat degradation', 'alien invasion', 'land use change', 'development activities', 'unsustainable fishing', 'destructive fishing', 'pollution' and 'water abstraction', which will be useful to prioritize strategic policy planning for NIFF resources. Though designed for India, its reproducibility and scalability will make it valuable for other countries concerning fish diversity and conservation research.