Benjamin_Boenisch/breakpilot-compliance
1
1
Fork 0
Code Issues 24 Pull Requests Actions Packages Projects Releases Wiki Activity

test(ai-sdk): GT #3 (commercial dishwasher) + fix Drehtisch keyword mislabel

Browse Source 
Add ground_truth_warewashing.json + TestWarewashing_GTCoverage. The test runs
the UC-M narrative through the SAME chain as production (ParseNarrative ->
engine -> relevance + cyber filter), so keyword/gating fixes are measured on
the real hazard set, and false positives show up as "extra".

Class A (generic keyword hygiene): spuelarm/spuelfeld no longer map to library
component C004 ("Drehtisch" / rotary table) — that mislabelled the spray arm.
Keep the rotating_part tag. Removes the bogus "Drehtisch" hazard.

GT #3 baseline -> after Class A: recall 80% (unchanged), one false positive
(Drehtisch) removed. Kistenhub 97.1% and Bremse pinned mappings unchanged.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-06-24 21:51:26 +02:00
parent 79ad95e244
commit cf86dc241b
3 changed files with 384 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ai-compliance-sdk/internal/iace/keyword_dictionary.go
+5 -1
View File
@@ -101,7 +101,11 @@ func GetKeywordDictionary() []KeywordEntry {
{Keywords: []string{"dampf", "wrasen", "schwaden", "brueden"}, ExtraTags: []string{"steam_emission", "high_temperature"}},
{Keywords: []string{"boiler", "spuelboiler", "nachspuelboiler", "tankheiz", "boilerheiz"}, ComponentIDs: []string{"C094"}, ExtraTags: []string{"heating_element", "high_temperature"}},
{Keywords: []string{"reiniger", "klarspueler", "spuelmittel", "reinigungsmittel", "reinigerkonzentrat", "spuelchemie", "dosiergeraet", "dosierpumpe", "sauglanze", "entkalker"}, ExtraTags: []string{"corrosive_chemical"}},
{Keywords: []string{"spuelarm", "spuelfeld", "wascharm", "spruehfeld"}, ComponentIDs: []string{"C004"}, ExtraTags: []string{"rotating_part"}},
// Spuelarm/Spuelfeld emit only the rotating_part capability tag. They are
// NOT mapped to a library component — C004 is a "Drehtisch" (rotary table)
// and that mislabels the spray arm. Keyword->component must be semantically
// honest (generic hygiene; surfaced by the warewashing GT).
{Keywords: []string{"spuelarm", "spuelfeld", "wascharm", "spruehfeld"}, ExtraTags: []string{"rotating_part"}},
{Keywords: []string{"spuelkammer", "spueltuer", "geraetetuer", "haubentuer", "klapptuer"}, ExtraTags: []string{"access_door"}},
// Ghost-Closure (Emit-Seite): macht die 34 toten Required-Tags
// emittierbar, jeweils NUR via domaenenspezifische Keywords -> die 120