docs: distinguish change-certainty from classifier confidence (#75) - #86
Merged
Conversation
The 0.65 MACHINE_CONFIDENCE_CAP applies to classifier-scored discoveries. Direct-document change detections on tracked records store relevanceScore 1 by design — the score records certainty that a known instrument's page changed. All four doc sites now say so instead of overclaiming a universal cap. Both kinds of detection remain editor-gated. Closes #75 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
MACHINE_CONFIDENCE_CAP) applies to classifier-scored discoveries only; the direct-document fingerprint path storesrelevanceScore: 1by design — change-certainty for an already-tracked instrument, not classifier confidence.Finding 2 (dedup) — investigated, no defect
The 7 Aug double emission for
qld-qgea-aiwas not a dedup miss. Watch-state history shows the page hash went A→B→A: baseline9a4acb…(29 Jul), the final Actions run retrieved8c9197…at 00:47 UTC (seq 2 →…p86jgk), then the first Argus run retrieved9a4acb…again at 03:01 UTC (seq 3 →…1oub5xk). Two distinct hash transitions, each correctly emitted once; the documented guarantee ("a version that reappears after intervening versions is reviewed again") describes exactly this case. One-time cutover artifact of both collectors running the same day from different environments.Closes #75
🤖 Generated with Claude Code