{
  "id": "specs/decisions/inference/minhash-for-twin-resources",
  "mentions": [
    {
      "context": "decisions",
      "file": {
        "href": "../../../../briefs/meetings/2026/2026-07-09-twin-resources-review/twin-resources-review/index.html",
        "label": "meetings/2026/2026-07-09-twin-resources-review/twin-resources-review.md"
      },
      "href": "../../../../briefs/meetings/2026/2026-07-09-twin-resources-review/twin-resources-review/index.html#L1",
      "kind": "written",
      "line": 1,
      "passages": 3,
      "title": "Twin resources review",
      "type": "meeting",
      "typeLabel": "Meeting"
    },
    {
      "context": "…e and its content signal is capped. Resources are sorted by identifier before an…",
      "file": {
        "href": "../../../objects/ingestion/resource/index.html",
        "label": "objects/ingestion/resource.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "passages": 3,
      "surface": "Resources",
      "title": "Resource",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "Representation",
      "file": {
        "href": "../../../objects/ingestion/representation/index.html",
        "label": "objects/ingestion/representation.md"
      },
      "href": "#L13",
      "kind": "written",
      "line": 13,
      "passages": 2,
      "surface": "Representation",
      "title": "Representation",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "Duplicate candidate",
      "file": {
        "href": "../../../rules/documents/duplicate-candidate/index.html",
        "label": "rules/documents/duplicate-candidate.rule.md"
      },
      "href": "#L14",
      "kind": "written",
      "line": 14,
      "passages": 2,
      "surface": "Duplicate candidate",
      "title": "Duplicate candidate",
      "type": "rule",
      "typeLabel": "Business rule"
    },
    {
      "context": "…very pair. Each text goes to its comparison form, then to shingles of five words…",
      "file": {
        "href": "../../../../glossary/inference/recognition/comparison-form/index.html",
        "label": "inference/recognition/comparison-form.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "comparison form",
      "title": "Comparison form",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "…at helps recognise the forms of one document works on extracted text, never on b…",
      "file": {
        "href": "../../../../glossary/ingestion/documents/document/index.html",
        "label": "ingestion/documents/document.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "document",
      "title": "Document",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "MinHash for twin resources: the estimate on 128 functions decides which pairs ar…",
      "file": {
        "href": "../../../../briefs/meetings/2026/2026-07-09-twin-resources-review/twin-resources-review/index.html",
        "label": "meetings/2026/2026-07-09-twin-resources-review/twin-resources-review.md"
      },
      "href": "../../../../briefs/meetings/2026/2026-07-09-twin-resources-review/twin-resources-review/index.html#L21",
      "kind": "recognised",
      "line": 21,
      "passages": 3,
      "surface": "MinHash for twin resources",
      "title": "Twin resources review",
      "type": "meeting",
      "typeLabel": "Meeting"
    },
    {
      "context": "…sion of the session, then: minhash for twin resources, as Participant-2 describe…",
      "file": {
        "href": "../../../../briefs/meetings/2026/2026-07-09-twin-resources-review/twin-resources-review/index.html",
        "label": "meetings/2026/2026-07-09-twin-resources-review/twin-resources-review.vtt"
      },
      "href": "../../../../briefs/meetings/2026/2026-07-09-twin-resources-review/twin-resources-review/index.html#L14",
      "kind": "recognised",
      "line": 14,
      "location": "00:57:55",
      "passages": 3,
      "surface": "minhash for twin resources",
      "title": "Twin resources review",
      "type": "meeting",
      "typeLabel": "Meeting"
    },
    {
      "context": "…e and its content signal is capped. Resources are sorted by identifier before an…",
      "file": {
        "href": "../../../objects/ingestion/resource/index.html",
        "label": "objects/ingestion/resource.md"
      },
      "href": "#L12",
      "kind": "written",
      "line": 12,
      "passages": 3,
      "surface": "Resource",
      "title": "Resource",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "Resource",
      "file": {
        "href": "../../../objects/ingestion/resource/index.html",
        "label": "objects/ingestion/resource.md"
      },
      "href": "#L12",
      "kind": "recognised",
      "line": 12,
      "passages": 3,
      "surface": "Resource",
      "title": "Resource",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "Representation",
      "file": {
        "href": "../../../objects/ingestion/representation/index.html",
        "label": "objects/ingestion/representation.md"
      },
      "href": "#L13",
      "kind": "recognised",
      "line": 13,
      "passages": 2,
      "surface": "Representation",
      "title": "Representation",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "Duplicate candidate",
      "file": {
        "href": "../../../rules/documents/duplicate-candidate/index.html",
        "label": "rules/documents/duplicate-candidate.rule.md"
      },
      "href": "#L14",
      "kind": "recognised",
      "line": 14,
      "passages": 2,
      "surface": "Duplicate candidate",
      "title": "Duplicate candidate",
      "type": "rule",
      "typeLabel": "Business rule"
    },
    {
      "context": "…e forms of one document works on extracted text, never on binary content, and ne…",
      "file": {
        "href": "../../../../glossary/ingestion/documents/extracted-text/index.html",
        "label": "ingestion/documents/extracted-text.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "extracted text",
      "title": "Extracted text",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "…is capped. Resources are sorted by identifier before anything else, so two runs …",
      "file": {
        "href": "../../../../glossary/ingestion/typing/identifier/index.html",
        "label": "ingestion/typing/identifier.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "identifier",
      "title": "Identifier",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "… 0.5 or more have their exact Jaccard index recomputed on the full shingle sets,…",
      "file": {
        "href": "../../../../glossary/publication/search/alphabetical-index/index.html",
        "label": "publication/search/alphabetical-index.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "index",
      "title": "Alphabetical index",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "… 0.5 or more have their exact Jaccard index recomputed on the full shingle sets,…",
      "file": {
        "href": "../../../../glossary/publication/search/search-index/index.html",
        "label": "publication/search/search-index.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "index",
      "title": "Search index",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "…ed on the full shingle sets, and the finding gives the share of lines in common;…",
      "file": {
        "href": "../../../../glossary/quality/findings/finding/index.html",
        "label": "quality/findings/finding.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "finding",
      "title": "Finding",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "… four rows per band, enumerates the candidate pairs, two signatures that share n…",
      "file": {
        "href": "../../../objects/inference/candidate/index.html",
        "label": "objects/inference/candidate.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "candidate",
      "title": "Candidate",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "…ed on the full shingle sets, and the finding gives the share of lines in common;…",
      "file": {
        "href": "../../../objects/quality/finding/index.html",
        "label": "objects/quality/finding.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "finding",
      "title": "Finding",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "… 0.5 or more have their exact Jaccard index recomputed on the full shingle sets,…",
      "file": {
        "href": "../../../screens/pages/alphabetical-index/index.html",
        "label": "screens/pages/alphabetical-index.md"
      },
      "href": "#L8",
      "kind": "recognised",
      "line": 8,
      "surface": "index",
      "title": "Alphabetical index",
      "type": "screen",
      "typeLabel": "Screen"
    },
    {
      "context": "Representation",
      "file": {
        "href": "../../../../glossary/ingestion/sources/representation/index.html",
        "label": "ingestion/sources/representation.md"
      },
      "href": "#L13",
      "kind": "recognised",
      "line": 13,
      "surface": "Representation",
      "title": "Representation",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "Build",
      "file": {
        "href": "../../../../glossary/ingestion/pipeline/build/index.html",
        "label": "ingestion/pipeline/build.md"
      },
      "href": "#L15",
      "kind": "recognised",
      "line": 15,
      "surface": "Build",
      "title": "Build",
      "type": "term",
      "typeLabel": "Term"
    },
    {
      "context": "Build",
      "file": {
        "href": "../../../objects/ingestion/build/index.html",
        "label": "objects/ingestion/build.md"
      },
      "href": "#L15",
      "kind": "recognised",
      "line": 15,
      "surface": "Build",
      "title": "Build",
      "type": "business_object",
      "typeLabel": "Business object"
    },
    {
      "context": "Build",
      "file": {
        "href": "../../../processes/ingestion/build-pipeline/index.html",
        "label": "processes/ingestion/build-pipeline.md"
      },
      "href": "#L15",
      "kind": "written",
      "line": 15,
      "surface": "Build",
      "title": "Build pipeline",
      "type": "process",
      "typeLabel": "Process"
    },
    {
      "context": "…e end of the row when the twin-resource reconciliation merged them. A deck and t…",
      "file": {
        "href": "../../../screens/pages/meeting-page/index.html",
        "label": "screens/pages/meeting-page.md"
      },
      "href": "../../../screens/pages/meeting-page/index.html#L16",
      "kind": "written",
      "line": 16,
      "surface": "twin-resource reconciliation",
      "title": "Meeting page",
      "type": "screen",
      "typeLabel": "Screen"
    }
  ]
}