Evidence-Linked Machine Learning Data Package for Prioritizing Fluorinated Organic Contaminants

This dataset supports the manuscript “Evidence-Linked Machine Learning Prioritizes Fluorinated Organic Contaminants for Monitoring and Transformation Screening”. The archive contains matched and curated public data tables, model-ready inputs, machine-learning outputs, validation summaries, uncertainty and transferability analyses, precursor-motif screening results, priority-ranking tables, and figure/table source data used in the study. The deposited files include raw_public_data and project_results_data directories with file manifests, row counts, and SHA-256 checksums. The dataset deposits the matched, curated, and analysis-ready public records used in the study, together with provenance manifests for the complete public database archives consulted. It does not redistribute full third-party database archives unless otherwise indicated.