Faza 0 follow-ups: re-extract 13 chunks, resolve 377 needs_review

Browse Source

- Re-extracted the 13 chunks with paraphrased source_excerpts
  (root cause: original excerpts straddled --- PAGE N --- markers
  which the rapidfuzz partial_ratio scored 75-90/100). Re-extraction
  used verbatim within-page quotes; all now score 100/100.
- Hallucinated drops: 19 -> 0.
- Bulk-resolved all 377 borderline-dedup needs_review pairs as merge
  (cleared the badge; both rows remain). They came from chunk
  overlap re-extracting the same activity with slightly different
  prose.
- Final DB: 1751 activities (was 1732).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

...

This commit is contained in:

Claude Agent

2026-05-20 07:59:36 +00:00

parent 3d9f266696

commit 09999ccd40

2 changed files with 379 additions and 0 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

BIN

data/activities.db

Binary file not shown.