Classify

The classify node sends each document page to a specific output branch based on its content. Use it to sort incoming documents by type, topic, or any other classification criteria before applying downstream processing.

​

When to use classify

• You receive mixed document types through one trigger and need to apply different processing to each.
• Different document types call for different extract schemas. Classifying upstream lets each branch use a focused schema instead of one mega-schema covering everything.
• You want to drop or quarantine documents that don’t fit any expected type. Use the other output for unmatched pages.
• Use route instead when you’re filtering on a metadata condition that doesn’t require reading the document content.

For mode selection, criteria writing, and confidence tuning, see classify best practices.

​

Configuration

​

Modes

​

Rules

Define conditions that match page content or page numbers. Each rule maps to a different output edge. Rules accept Contains text, Not contains text, and Page numbers.

​

AI

Uses AI embedding similarity to match pages against reference documents in your target pipelines. Select one or more Target Pipelines. Each target pipeline must have a Forward Trigger with reference documents so Ingestly can compare incoming pages against it. Set the Confidence Threshold to control how strict the match must be. Pages with a similarity score at or above the threshold are forwarded to the matching pipeline; pages below fall through to the other output so you can handle them separately.

​

Inputs and outputs

Allowed inputs: Trigger nodes, route, parse, review. Output: Pages routed to the matching output edge, or to the other edge if no classification matched.

​

Credits

See credits for pricing. Rules mode costs 0.5 credits per page; AI mode costs 0.2 credits per page. Generating embeddings on the forward trigger when you upload reference documents is billed separately at 0.2 credits per page per reference document.

​

Related