Interface: IngestionConfigDto
@kortexya/reasoninglayer / Ingestion / IngestionConfigDto
Interface: IngestionConfigDto
Defined in: src/types/ingestion.ts:376
Configuration for ingestion.
Remarks
Controls LLM extraction, NER, deduplication, schema discovery, and various quality pipeline options.
Properties
autoCreateSorts?
optionalautoCreateSorts:boolean
Defined in: src/types/ingestion.ts:378
Whether to automatically create new sorts for unrecognized entity types.
batchSize?
optionalbatchSize:number
Defined in: src/types/ingestion.ts:380
Batch size for parallel processing.
deduplicationThreshold?
optionaldeduplicationThreshold:number
Defined in: src/types/ingestion.ts:382
Confidence threshold for deduplication (0.0 - 1.0).
documentId?
optionaldocumentId:string|null
Defined in: src/types/ingestion.ts:384
Optional document ID for tracking sort provenance.
domainInstructions?
optionaldomainInstructions:string[]
Defined in: src/types/ingestion.ts:386
Domain-specific instructions for LLM extraction.
enableDescriptionConsolidation?
optionalenableDescriptionConsolidation:boolean
Defined in: src/types/ingestion.ts:388
Enable cross-chunk description consolidation. Default: true.
enableEntityFiltering?
optionalenableEntityFiltering:boolean
Defined in: src/types/ingestion.ts:390
Enable post-extraction entity filtering. Default: true.
enableEvidenceEnrichment?
optionalenableEvidenceEnrichment:boolean
Defined in: src/types/ingestion.ts:396
Enable LLM-based evidence enrichment with statistical features. Default: false.
enableFocusedExtraction?
optionalenableFocusedExtraction:boolean
Defined in: src/types/ingestion.ts:392
Enable focused extraction prompts. Default: true.
enableSchemaDiscovery?
optionalenableSchemaDiscovery:boolean
Defined in: src/types/ingestion.ts:394
Enable schema discovery before extraction. Default: true.
enableZeroDegreeFiltering?
optionalenableZeroDegreeFiltering:boolean
Defined in: src/types/ingestion.ts:398
Enable zero-degree entity removal. Default: false.
evidenceDerivationConfig?
optionalevidenceDerivationConfig:EvidenceDerivationConfigDto|null
Defined in: src/types/ingestion.ts:403
Configuration for deriving evidence terms from relational features. When set, entities with Reference features automatically get linked evidence terms.
extractionStrategy?
optionalextractionStrategy:ExtractionStrategy
Defined in: src/types/ingestion.ts:418
Extraction strategy — an internally tagged discriminated union.
Remarks
Wire format uses #[serde(tag = "type")]. Pass an object with a type discriminator.
Example
{ type: 'llm' }{ type: 'hybrid', nerConfidenceThreshold: 0.8 }extractRelations?
optionalextractRelations:boolean
Defined in: src/types/ingestion.ts:405
Whether to extract relations between entities.
forceExtractionStrategy?
optionalforceExtractionStrategy:boolean
Defined in: src/types/ingestion.ts:423
When true, skip the auto-upgrade from Llm to LocalNer even if NER is available. Used for A/B comparison tests. Default: false.
maxSchemaTokens?
optionalmaxSchemaTokens:number
Defined in: src/types/ingestion.ts:425
Maximum tokens for schema context passed to LLM.
nerConfidenceThreshold?
optionalnerConfidenceThreshold:number
Defined in: src/types/ingestion.ts:427
NER confidence threshold (0.0 - 1.0). Default: 0.5.
nerEntityLabels?
optionalnerEntityLabels:string[]
Defined in: src/types/ingestion.ts:429
Entity labels for NER extraction.
nerRelationConfidenceThreshold?
optionalnerRelationConfidenceThreshold:number
Defined in: src/types/ingestion.ts:431
NER relation confidence threshold (0.0 - 1.0). Default: 0.5.
nerRelationLabels?
optionalnerRelationLabels:string[]
Defined in: src/types/ingestion.ts:433
Relation labels for NER relation extraction.
rejectAnonymizedReferences?
optionalrejectAnonymizedReferences:boolean
Defined in: src/types/ingestion.ts:439
Reject anonymized reference patterns from case studies. Filters patterns like “Monsieur X”, “société A”, “Company A”. Only active when entity filtering is enabled. Default: true.
skipFingerprint?
optionalskipFingerprint:boolean
Defined in: src/types/ingestion.ts:441
Skip fingerprint check (force re-ingestion).
skipRelationCompletion?
optionalskipRelationCompletion:boolean
Defined in: src/types/ingestion.ts:443
Skip automatic relation completion between extracted entities.