Field-level data provenance | Synthetic demo

Self-directed synthetic demo, created with AI assistance. All source notes and organizations are fictional. No websites were researched, and no paid-client result is claimed.

This project structures supplied fictional business notes into a 12-organization dataset with 36 field-level evidence records. Explicit mappings standardize 34 fields; two ambiguous or missing fields stay unresolved. Original phrases and source IDs are preserved.

The example.com URLs are placeholders. The work demonstrates structured data preparation, taxonomy mapping, provenance and uncertainty handling rather than verified web research.

The project includes a workbook, original source CSVs, prepared data and reproducible quality checks. Actual demo workbook previews are shown. Every resolved field has a recorded basis, and uncertainty remains visible for review.