Member Junction
    Preparing search index...
    interface DuplicateEntryCheckOptions {
        AbsoluteMatchThreshold?: number;
        CancellationToken?: AbortSignal;
        DuplicateRunID?: string;
        FusionMethod?: "rrf" | "weighted";
        IncrementalOnly?: boolean;
        KeywordSearchWeight?: number;
        OnProgress?: (progress: DuplicateDetectionProgress) => void;
        PotentialMatchThreshold?: number;
        ReRankingEnabled?: boolean;
        ReRankingModelID?: string;
        ReRankingTopK?: number;
        Revectorize?: boolean;
        TimeoutMS?: number;
        TopK?: number;
    }

    Hierarchy (View Summary)

    Index

    Properties

    AbsoluteMatchThreshold?: number

    Override the entity document's AbsoluteMatchThreshold for this run. Value between 0 and 1. If omitted, uses the entity document's value.

    CancellationToken?: AbortSignal

    Stops the check when aborted. A stopped check is Failed and flags nothing.

    DuplicateRunID?: string

    ID of an existing Duplicate Run record to continue

    FusionMethod?: "rrf" | "weighted"

    Fusion method when combining vector + keyword results (default: 'rrf')

    IncrementalOnly?: boolean

    Enable incremental mode — only check records not in a completed prior run (default: false)

    KeywordSearchWeight?: number

    Weight for keyword search in hybrid mode: 0.0 = pure vector, 1.0 = pure keyword (default: 0.3)

    OnProgress?: (progress: DuplicateDetectionProgress) => void

    Progress callback invoked at natural milestones during detection

    PotentialMatchThreshold?: number

    Override the entity document's PotentialMatchThreshold for this run. Value between 0 and 1 (e.g., 0.30 = 30%). If omitted, uses the entity document's value.

    ReRankingEnabled?: boolean

    Enable post-retrieval reranking via BaseReranker (default: false)

    ReRankingModelID?: string

    AI Model ID for the reranker; if omitted, uses default reranker

    ReRankingTopK?: number

    Max candidates to send to reranker per record (default: all retrieved)

    Revectorize?: boolean

    Re-vectorize records before detection (default: false). When false, assumes vectors already exist in the index from a prior sync. Set to true to force a fresh vectorization pass before running detection.

    TimeoutMS?: number

    The check's time budget, in milliseconds. Defaults to DUPLICATE_ENTRY_CHECK_SERVER_BUDGET_MS. It bounds every step: the template, the embedding, the vector query, both RunViews and the decision call.

    TopK?: number

    Number of nearest neighbors to retrieve per record (default: 5)