glam/data/rag_eval/golden_dataset.json
2026-01-09 20:35:19 +01:00

860 lines
24 KiB
JSON

{
"version": "1.0.0",
"description": "Golden dataset for DSPy RAG evaluation - Dutch heritage institutions COUNT queries",
"created_at": "2025-01-09",
"metadata": {
"source": "Oxigraph knowledge graph SPARQL queries",
"institution_counts_verified_at": "2025-01-08",
"total_examples": 60
},
"examples": [
{
"id": "count_001",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Utrecht?",
"expected_count": 10,
"expected_slots": {
"institution_type": "A",
"location": "NL-UT",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_002",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Noord-Holland?",
"expected_count": 26,
"expected_slots": {
"institution_type": "A",
"location": "NL-NH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_003",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Zuid-Holland?",
"expected_count": 35,
"expected_slots": {
"institution_type": "A",
"location": "NL-ZH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_004",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Gelderland?",
"expected_count": 28,
"expected_slots": {
"institution_type": "A",
"location": "NL-GE",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_005",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Noord-Brabant?",
"expected_count": 14,
"expected_slots": {
"institution_type": "A",
"location": "NL-NB",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_006",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Overijssel?",
"expected_count": 31,
"expected_slots": {
"institution_type": "A",
"location": "NL-OV",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_007",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Friesland?",
"expected_count": 24,
"expected_slots": {
"institution_type": "A",
"location": "NL-FR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_008",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Limburg?",
"expected_count": 15,
"expected_slots": {
"institution_type": "A",
"location": "NL-LI",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_009",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Zeeland?",
"expected_count": 8,
"expected_slots": {
"institution_type": "A",
"location": "NL-ZE",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_010",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Drenthe?",
"expected_count": 13,
"expected_slots": {
"institution_type": "A",
"location": "NL-DR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_011",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Groningen?",
"expected_count": 28,
"expected_slots": {
"institution_type": "A",
"location": "NL-GR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_012",
"category": "count",
"subcategory": "archive_by_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in Flevoland?",
"expected_count": 3,
"expected_slots": {
"institution_type": "A",
"location": "NL-FL",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_013",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Utrecht?",
"expected_count": 59,
"expected_slots": {
"institution_type": "M",
"location": "NL-UT",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_014",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Noord-Holland?",
"expected_count": 204,
"expected_slots": {
"institution_type": "M",
"location": "NL-NH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_015",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Zuid-Holland?",
"expected_count": 164,
"expected_slots": {
"institution_type": "M",
"location": "NL-ZH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_016",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Gelderland?",
"expected_count": 135,
"expected_slots": {
"institution_type": "M",
"location": "NL-GE",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_017",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Noord-Brabant?",
"expected_count": 114,
"expected_slots": {
"institution_type": "M",
"location": "NL-NB",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_018",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Overijssel?",
"expected_count": 80,
"expected_slots": {
"institution_type": "M",
"location": "NL-OV",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_019",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Friesland?",
"expected_count": 70,
"expected_slots": {
"institution_type": "M",
"location": "NL-FR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_020",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Limburg?",
"expected_count": 53,
"expected_slots": {
"institution_type": "M",
"location": "NL-LI",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_021",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Zeeland?",
"expected_count": 48,
"expected_slots": {
"institution_type": "M",
"location": "NL-ZE",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_022",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Drenthe?",
"expected_count": 32,
"expected_slots": {
"institution_type": "M",
"location": "NL-DR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_023",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Groningen?",
"expected_count": 31,
"expected_slots": {
"institution_type": "M",
"location": "NL-GR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_024",
"category": "count",
"subcategory": "museum_by_province",
"language": "nl",
"question": "Hoeveel musea zijn er in Flevoland?",
"expected_count": 8,
"expected_slots": {
"institution_type": "M",
"location": "NL-FL",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_025",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Utrecht?",
"expected_count": 13,
"expected_slots": {
"institution_type": "L",
"location": "NL-UT",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_026",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Noord-Holland?",
"expected_count": 25,
"expected_slots": {
"institution_type": "L",
"location": "NL-NH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_027",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Zuid-Holland?",
"expected_count": 39,
"expected_slots": {
"institution_type": "L",
"location": "NL-ZH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_028",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Gelderland?",
"expected_count": 23,
"expected_slots": {
"institution_type": "L",
"location": "NL-GE",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_029",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Noord-Brabant?",
"expected_count": 24,
"expected_slots": {
"institution_type": "L",
"location": "NL-NB",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_030",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Overijssel?",
"expected_count": 32,
"expected_slots": {
"institution_type": "L",
"location": "NL-OV",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_031",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Friesland?",
"expected_count": 6,
"expected_slots": {
"institution_type": "L",
"location": "NL-FR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_032",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Limburg?",
"expected_count": 17,
"expected_slots": {
"institution_type": "L",
"location": "NL-LI",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_033",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Zeeland?",
"expected_count": 8,
"expected_slots": {
"institution_type": "L",
"location": "NL-ZE",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_034",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Drenthe?",
"expected_count": 5,
"expected_slots": {
"institution_type": "L",
"location": "NL-DR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_035",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Groningen?",
"expected_count": 5,
"expected_slots": {
"institution_type": "L",
"location": "NL-GR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_036",
"category": "count",
"subcategory": "library_by_province",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Flevoland?",
"expected_count": 5,
"expected_slots": {
"institution_type": "L",
"location": "NL-FL",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_037",
"category": "count",
"subcategory": "alternative_phrasing",
"language": "nl",
"question": "Wat is het aantal archieven in Utrecht?",
"expected_count": 10,
"expected_slots": {
"institution_type": "A",
"location": "NL-UT",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_038",
"category": "count",
"subcategory": "alternative_phrasing",
"language": "nl",
"question": "Kun je me vertellen hoeveel musea er in Amsterdam zijn?",
"expected_count": 75,
"expected_slots": {
"institution_type": "M",
"location": "Amsterdam",
"location_level": "settlement",
"response_mode": "count"
},
"notes": "City-level query - Amsterdam museums"
},
{
"id": "count_039",
"category": "count",
"subcategory": "alternative_phrasing",
"language": "nl",
"question": "Tel het aantal bibliotheken in Zuid-Holland",
"expected_count": 39,
"expected_slots": {
"institution_type": "L",
"location": "NL-ZH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_040",
"category": "count",
"subcategory": "alternative_phrasing",
"language": "nl",
"question": "Geef het totaal aantal archieven in Noord-Holland",
"expected_count": 26,
"expected_slots": {
"institution_type": "A",
"location": "NL-NH",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_041",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel archieven zijn er in Den Haag?",
"expected_count": 8,
"expected_slots": {
"institution_type": "A",
"location": "Den Haag",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_042",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel musea zijn er in Rotterdam?",
"expected_count": 35,
"expected_slots": {
"institution_type": "M",
"location": "Rotterdam",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_043",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Utrecht stad?",
"expected_count": 6,
"expected_slots": {
"institution_type": "L",
"location": "Utrecht",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_044",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel archieven heeft Haarlem?",
"expected_count": 3,
"expected_slots": {
"institution_type": "A",
"location": "Haarlem",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_045",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel musea staan er in Leiden?",
"expected_count": 18,
"expected_slots": {
"institution_type": "M",
"location": "Leiden",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_046",
"category": "count",
"subcategory": "informal_query",
"language": "nl",
"question": "Zijn er veel archieven in Zeeland?",
"expected_count": 8,
"expected_slots": {
"institution_type": "A",
"location": "NL-ZE",
"location_level": "subregion",
"response_mode": "count"
},
"notes": "Informal phrasing - should still return count"
},
{
"id": "count_047",
"category": "count",
"subcategory": "informal_query",
"language": "nl",
"question": "Welke musea zijn er in Flevoland en hoeveel?",
"expected_count": 8,
"expected_slots": {
"institution_type": "M",
"location": "NL-FL",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_048",
"category": "count",
"subcategory": "informal_query",
"language": "nl",
"question": "Ik zoek archieven in Drenthe, hoeveel zijn er?",
"expected_count": 13,
"expected_slots": {
"institution_type": "A",
"location": "NL-DR",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_049",
"category": "count",
"subcategory": "edge_case",
"language": "nl",
"question": "Hoeveel archieven en musea zijn er in Groningen?",
"expected_count": null,
"expected_slots": {
"institution_type": ["A", "M"],
"location": "NL-GR",
"location_level": "subregion",
"response_mode": "count"
},
"notes": "Multiple institution types - expected to be split or summed (28+31=59)"
},
{
"id": "count_050",
"category": "count",
"subcategory": "edge_case",
"language": "nl",
"question": "Hoeveel erfgoedinstellingen zijn er in Limburg?",
"expected_count": null,
"expected_slots": {
"institution_type": "*",
"location": "NL-LI",
"location_level": "subregion",
"response_mode": "count"
},
"notes": "Generic 'heritage institutions' - should count all types (15+53+17=85)"
},
{
"id": "count_051",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel archieven zijn er in Maastricht?",
"expected_count": 5,
"expected_slots": {
"institution_type": "A",
"location": "Maastricht",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_052",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel musea zijn er in Arnhem?",
"expected_count": 12,
"expected_slots": {
"institution_type": "M",
"location": "Arnhem",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_053",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel archieven zijn er in Eindhoven?",
"expected_count": 3,
"expected_slots": {
"institution_type": "A",
"location": "Eindhoven",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_054",
"category": "count",
"subcategory": "city_level",
"language": "nl",
"question": "Hoeveel bibliotheken zijn er in Groningen stad?",
"expected_count": 4,
"expected_slots": {
"institution_type": "L",
"location": "Groningen",
"location_level": "settlement",
"response_mode": "count"
}
},
{
"id": "count_055",
"category": "count",
"subcategory": "formal_dutch",
"language": "nl",
"question": "Kunt u mij informeren over het aantal archieven in de provincie Utrecht?",
"expected_count": 10,
"expected_slots": {
"institution_type": "A",
"location": "NL-UT",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_056",
"category": "count",
"subcategory": "formal_dutch",
"language": "nl",
"question": "Graag zou ik willen weten hoeveel musea er in Noord-Brabant gevestigd zijn",
"expected_count": 114,
"expected_slots": {
"institution_type": "M",
"location": "NL-NB",
"location_level": "subregion",
"response_mode": "count"
}
},
{
"id": "count_057",
"category": "count",
"subcategory": "abbreviated_province",
"language": "nl",
"question": "Hoeveel archieven zijn er in NH?",
"expected_count": 26,
"expected_slots": {
"institution_type": "A",
"location": "NL-NH",
"location_level": "subregion",
"response_mode": "count"
},
"notes": "Abbreviated province name"
},
{
"id": "count_058",
"category": "count",
"subcategory": "abbreviated_province",
"language": "nl",
"question": "Hoeveel musea in ZH?",
"expected_count": 164,
"expected_slots": {
"institution_type": "M",
"location": "NL-ZH",
"location_level": "subregion",
"response_mode": "count"
},
"notes": "Very short query with abbreviated province"
},
{
"id": "count_059",
"category": "count",
"subcategory": "misspelling",
"language": "nl",
"question": "Hoeveel archiefen zijn er in Utrecht?",
"expected_count": 10,
"expected_slots": {
"institution_type": "A",
"location": "NL-UT",
"location_level": "subregion",
"response_mode": "count"
},
"notes": "Common misspelling of 'archieven'"
},
{
"id": "count_060",
"category": "count",
"subcategory": "misspelling",
"language": "nl",
"question": "Hoeveel musea zijn er in Noordholland?",
"expected_count": 204,
"expected_slots": {
"institution_type": "M",
"location": "NL-NH",
"location_level": "subregion",
"response_mode": "count"
},
"notes": "Missing hyphen in 'Noord-Holland'"
}
]
}