Inferensys

Glossary

DefinedTerm

A Schema.org structured data type used to markup a single word or phrase and its formal definition within a DefinedTermSet, enabling search engines to understand and display glossary-style content.
Developer reviewing semantic search engine results on laptop, relevance scores visible, technical search demo.
SCHEMA.ORG VOCABULARY

What is DefinedTerm?

A Schema.org type for formally defining a single word, phrase, or acronym within a structured glossary, enabling search engines to parse and display precise definitions in knowledge panels and rich results.

DefinedTerm is a Schema.org structured data type used to markup a single word or phrase and its corresponding formal definition, typically within a parent DefinedTermSet. By wrapping glossary entries in this schema, web architects provide search engines with an unambiguous, machine-readable signal that disambiguates the term's meaning and establishes the page as an authoritative lexical source for entity extraction.

The type requires a name property for the term and a description for its definition, while optional properties like termCode and inDefinedTermSet allow for precise entity reconciliation with external controlled vocabularies. When implemented via JSON-LD, DefinedTerm markup helps content qualify for glossary-rich features in search results, directly supporting Generative Engine Optimization by structuring knowledge for AI-driven answer engines.

SCHEMA.ORG TYPE

Key Properties of DefinedTerm

DefinedTerm is a Schema.org type that provides a formal, machine-readable definition for a single word or phrase within a DefinedTermSet. It enables search engines to understand glossary-style content and display rich definitions directly in search results.

01

Core Properties

The DefinedTerm type inherits from Intangible and includes several critical properties for semantic precision:

  • name: The word, phrase, or acronym being defined (e.g., "JSON-LD")
  • description: A formal, concise definition of the term
  • termCode: A machine-readable identifier, often an abbreviation or internal code
  • inDefinedTermSet: A reference to the parent DefinedTermSet that contains this term
  • url: A canonical link to the term's dedicated definition page
02

Entity Identification Properties

To disambiguate the term from other concepts with the same name, DefinedTerm supports standard entity reconciliation properties:

  • sameAs: Link to an external canonical URI, such as a Wikidata Q-ID or DBpedia resource, to explicitly assert the term's identity
  • identifier: A formal identifier from an external authority, such as a DOI or ISBN
  • image: A diagram or illustration that visually represents the concept

Using sameAs is critical for consolidating authority signals across the knowledge graph.

03

Relationship to DefinedTermSet

A DefinedTerm must belong to a parent DefinedTermSet via the inDefinedTermSet property. This container defines the scope and authority of the glossary:

  • The DefinedTermSet provides the name and description of the overall vocabulary
  • It establishes the topical domain (e.g., "AI Glossary" or "Legal Terms")
  • Search engines use this hierarchical relationship to understand that the term's definition is authoritative within a specific context

This parent-child structure mirrors how Taxonomy and Ontology systems organize controlled vocabularies.

04

Implementation with JSON-LD

DefinedTerm is implemented using JSON-LD, the recommended serialization format for Schema.org. A typical implementation embeds the term within a DefinedTermSet array:

  • The @type is set to "DefinedTerm"
  • The @context references https://schema.org
  • Each term is a discrete object within the hasDefinedTerm array of the parent set
  • The markup is placed in the <head> or <body> of the term's dedicated HTML page

This structured data qualifies content for glossary-rich results and enhances Entity Recognition by search crawlers.

05

Search Engine Display Triggers

When properly implemented, DefinedTerm markup can trigger enhanced search result displays:

  • Featured Snippets: Google may extract the description property to display a direct definition at the top of search results for "What is X?" queries
  • Knowledge Graph Panels: The term may be integrated into Google's Knowledge Graph as a distinct entity node
  • Glossary Carousels: Multiple terms from the same DefinedTermSet can appear in a horizontal carousel

This visibility is a core component of Generative Engine Optimization strategies, as AI overviews prioritize content with clear entity definitions.

06

Distinction from Other Schema Types

DefinedTerm is often confused with similar Schema.org types, but serves a distinct purpose:

  • Thing > Intangible > DefinedTerm: A formal glossary entry with a definition
  • Thing > Intangible > Enumeration: A fixed set of values, not individual definitions
  • Thing > CreativeWork > Article: A full-length piece of content, not a concise definition
  • Thing > Intangible > AlignmentObject: Describes educational alignment, not a term definition

Using the correct type ensures accurate Entity Disambiguation and prevents schema conflicts.

SCHEMA MARKUP ENGINEERING

Frequently Asked Questions

Clear, technical answers to the most common questions about implementing the DefinedTerm schema type for glossary-style content and entity disambiguation.

DefinedTerm is a Schema.org structured data type used to markup a single word, phrase, or acronym within a DefinedTermSet to provide a formal, machine-readable definition. It works by wrapping a glossary entry in JSON-LD or Microdata, explicitly linking the term (name) to its authoritative explanation (description) and, optionally, a canonical termCode. This enables search engines to parse the content not as ambiguous text, but as a discrete, disambiguated entity. When implemented correctly, a search engine can extract the definition directly for featured snippets, knowledge panels, or voice assistant responses, bypassing the need for probabilistic interpretation of the surrounding prose.

Prasad Kumkar

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.