ToolJovia
Data

Bounded Record Linkage Candidate Review

Generate cross-table text-similarity candidates inside explicit exact blocks while retaining ambiguity for human review.

✓ Browser-local processing

Loading tool…

How to use Bounded Record Linkage Candidate Review

Find records worth comparing without declaring two records to be the same person or entity. Exact block values control candidate generation, bounded Unicode-scalar edit distance supplies a transparent text measure, and multiple possible matches remain separate.

  1. Enter first csv, second csv, unique identifier column in both tables, comparison text column in both tables, exact blocking columns json (empty for all pairs), maximum edit distance (0–4).
  2. Choose the settings for bounded record linkage candidate review, then run the tool.
  3. Review the result and any limitations before copying or downloading it.

Good to know

  • 200 records per side, 100 Unicode scalars per compared value, 200 characters per ID, 10,000 planned pairs, 500,000 distance cells and 1,000 complete passing pairs. Browser computation runs in a module worker with a 2-second deadline after startup.
  • Exact blocks are not normalized; comparison text uses only declared ASCII case and outer whitespace policies. No accent removal, identity verification, probabilistic confidence, automatic merge, external records or ambiguous winner selection.
  • Processing is local. Output is limited to 1,000,000 text characters unless a stricter file limit is stated.

Common questions

Is the closest record automatically treated as a match?

No. Every passing pair remains a candidate for human review, including ties and multiple alternatives. Text similarity alone cannot establish identity or whether an absent candidate refers to a different entity.

Is this input sent to a server?

No. The application processes this input inside your browser tab and does not save it to persistent storage.

Keep going with related tools

All data →
Data

CSV Join Workbench

Join two CSV tables by explicit composite keys with inner, left or full record-preserving joins.

Text

String Similarity Calculator

Calculate grapheme Levenshtein edit distance and a clearly defined normalized similarity.

Data

Guarded CSV Cell Change Applier

Create a changed CSV only after every explicit key, target cell and expected prior value passes an atomic guard.