surface↓classification↓domain
Classificationdimension · returns varchar

DOMAIN

Detect subject matter domain for grouping (zero-shot NLI)

Per-row classifier — stable across GROUP BY.

classificationnlispecialist-zootext

Syntax

DOMAIN({{ text }})
{{ text }} BY DOMAIN

Arguments

nametypedescription
textVARCHAR—

About

Domain detection — per-row dimension that buckets text by subject matter domain. Useful for routing mixed-topic corpora. Backend: specialist zoo zero-shot NLI via /classify. Fixed label set of 13 common domains. Per-row call (no mapping mode) because domain classification doesn't benefit from batch-level consistency and the per-row path is simpler for a pure SELECT list use case. For LLM-style domain inference with more fine-grained or domain-specific taxonomies, use DOMAIN_LLM — see domain_dimension_llm.cascade.yaml.

Examples

Database content routed to a domain

SELECT
  domain ('SQL query optimization techniques')

Health content routed to a domain

SELECT
  domain ('Aspirin reduces inflammation')

Nearby rabbit holes

same domain
dimension

AUDIENCE

Identify target audience via zero-shot NLI

classificationnlispecialist-zootext+1
dimension

AUDIENCE_LLM

LLM-backed audience identification (escape hatch for AUDIENCE)

classificationllmllm-escape-hatchtext
dimension

AUTHENTICITY

Assess authenticity of content via zero-shot NLI

classificationnlispecialist-zootext
dimension

AUTHENTICITY_LLM

LLM-backed authenticity assessment (escape hatch for AUTHENTICITY)

classificationllmllm-escape-hatchtext
dimension

BUCKET

Classify text into user-specified buckets via zero-shot NLI

classificationnlispecialist-zootext
dimension

BUCKET_LLM

LLM-backed bucketing (escape hatch for BUCKET)

classificationllmllm-escape-hatchtext
Climb back to The Looking Glass