glam/scripts/batch12_final_validation.json
kempersc e5a532a8bc Add comprehensive tests for NLP institution extraction and RDF partnership integration
- Introduced `test_nlp_extractor.py` with unit tests for the InstitutionExtractor, covering various extraction patterns (ISIL, Wikidata, VIAF, city names) and ensuring proper classification of institutions (museum, library, archive).
- Added tests for extracted entities and result handling to validate the extraction process.
- Created `test_partnership_rdf_integration.py` to validate the end-to-end process of extracting partnerships from a conversation and exporting them to RDF format.
- Implemented tests for temporal properties in partnerships and ensured compliance with W3C Organization Ontology patterns.
- Verified that extracted partnerships are correctly linked with PROV-O provenance metadata.
2025-11-19 23:20:47 +01:00

42 lines
No EOL
1.3 KiB
JSON

{
"batch": 12,
"validation_date": "2025-11-09",
"institution_type": "LIBRARY",
"validated_matches": [],
"rejected_matches": [
{
"library_name": "Biblioteca Pública Federico Varela",
"city": "Chañaral",
"q_number": "Q135435755",
"wikidata_name": "Biblioteca Pública de La Reina",
"reason": "Location mismatch: Chañaral (Atacama) ≠ La Reina (Santiago)"
},
{
"library_name": "Biblioteca Pública de Illapel",
"city": "Los Vilos",
"q_number": "Q135435755",
"wikidata_name": "Biblioteca Pública de La Reina",
"reason": "Location mismatch: Los Vilos (Coquimbo) ≠ La Reina (Santiago)"
},
{
"library_name": "Biblioteca Pública N°56",
"city": "Talagante",
"q_number": "Q135435755",
"wikidata_name": "Biblioteca Pública de La Reina",
"reason": "Location mismatch: Talagante ≠ La Reina"
},
{
"library_name": "Biblioteca Pública N° 244",
"city": "Pichilemu",
"q_number": "Q134891536",
"wikidata_name": "Biblioteca Pública Municipal de Lebu",
"reason": "Location mismatch: Pichilemu (O'Higgins) ≠ Lebu (Bío Bío)"
}
],
"summary": {
"validated": 0,
"rejected": 4,
"false_positive_rate": "100.0%",
"recommendation": "REJECT ALL - Generic library names cause false positives"
}
}