For other uses, see [Knowledge graph (disambiguation)](https://en.wikipedia.org/wiki/Knowledge_graph_(disambiguation) "Knowledge graph (disambiguation)"). [](https://en.wikipedia.org/wiki/File:Conceptual_Diagram_-_Example.svg) Example conceptual diagram In [knowledge representation and reasoning](https://en.wikipedia.org/wiki/Knowledge_representation_and_reasoning "Knowledge representation and reasoning"), a **knowledge graph** is a [knowledge base](https://en.wikipedia.org/wiki/Knowledge_base "Knowledge base") that uses a [graph](https://en.wikipedia.org/wiki/Graph_(discrete_mathematics) "Graph (discrete mathematics)")-structured [data model](https://en.wikipedia.org/wiki/Data_model "Data model") or [topology](https://en.wikipedia.org/wiki/Topology "Topology") to represent and operate on [data](https://en.wikipedia.org/wiki/Data "Data"). Knowledge graphs are often used to store interlinked descriptions of [entities](https://en.wikipedia.org/wiki/Named_entity "Named entity") – objects, events, situations or abstract concepts – while also encoding the free-form [semantics](https://en.wikipedia.org/wiki/Semantics "Semantics") or relationships underlying these entities.[[1]](#cite_note-1)[[2]](#cite_note-2) Since the development of the [Semantic Web](https://en.wikipedia.org/wiki/Semantic_Web "Semantic Web"), knowledge graphs have often been associated with [linked open data](https://en.wikipedia.org/wiki/Linked_data "Linked data") projects, focusing on the connections between [concepts](https://en.wikipedia.org/wiki/Concept "Concept") and entities.[[3]](#cite_note-Ref1-3)[[4]](#cite_note-4) They are also historically associated with and used by [search engines](https://en.wikipedia.org/wiki/Search_engine "Search engine") such as [Google](https://en.wikipedia.org/wiki/Knowledge_Graph_(Google) "Knowledge Graph (Google)"), [Bing](https://en.wikipedia.org/wiki/Bing_(search_engine) "Bing (search engine)"), and [Yahoo](https://en.wikipedia.org/wiki/Yahoo "Yahoo"); [knowledge engines](https://en.wikipedia.org/wiki/Knowledge_engine "Knowledge engine") and question-answering services such as [WolframAlpha](https://en.wikipedia.org/wiki/WolframAlpha "WolframAlpha"), Apple's [Siri](https://en.wikipedia.org/wiki/Siri "Siri"), and [Amazon Alexa](https://en.wikipedia.org/wiki/Amazon_Alexa "Amazon Alexa"); and [social networks](https://en.wikipedia.org/wiki/Social_network "Social network") such as [LinkedIn](https://en.wikipedia.org/wiki/LinkedIn "LinkedIn") and [Facebook](https://en.wikipedia.org/wiki/Facebook "Facebook"). Recent developments in data science and [machine learning](https://en.wikipedia.org/wiki/Machine_learning "Machine learning"), particularly in [graph neural networks](https://en.wikipedia.org/wiki/Graph_neural_network "Graph neural network"), representation learning, and machine learning, have broadened the scope of knowledge graphs beyond their traditional use in search engines and [recommender systems](https://en.wikipedia.org/wiki/Recommender_system "Recommender system"). They are increasingly used in scientific research, with notable applications in fields such as [genomics](https://en.wikipedia.org/wiki/Genomics "Genomics"), [proteomics](https://en.wikipedia.org/wiki/Proteomics "Proteomics"), and [systems biology](https://en.wikipedia.org/wiki/Systems_biology "Systems biology").[[5]](#cite_note-5) History ------- [[edit](/w/index.php?title=Knowledge_graph&action=edit&section=1 "Edit section: History")] The term was coined as early as 1972 by the Austrian [linguist](https://en.wikipedia.org/wiki/Linguistics "Linguistics") [Edgar W. Schneider](https://en.wikipedia.org/wiki/Edgar_W._Schneider "Edgar W. Schneider"), in a discussion of how to build modular instructional systems for courses.[[6]](#cite_note-6) In the late 1980s, the [University of Groningen](https://en.wikipedia.org/wiki/University_of_Groningen "University of Groningen") and [University of Twente](https://en.wikipedia.org/wiki/University_of_Twente "University of Twente") jointly began a project called Knowledge Graphs, focusing on the design of [semantic networks](https://en.wikipedia.org/wiki/Semantic_network "Semantic network") with edges restricted to a limited set of relations, to facilitate [algebras on the graph](https://en.wikipedia.org/wiki/Graph_algebra "Graph algebra").[[7]](#cite_note-7) In subsequent decades, the distinction between semantic networks and knowledge graphs was blurred. Some early knowledge graphs were topic-specific. In 1985, [Wordnet](https://en.wikipedia.org/wiki/Wordnet "Wordnet") was founded, capturing semantic relationships between words and meanings – an application of this idea to language itself. In 2005, Marc Wirk founded [Geonames](https://en.wikipedia.org/wiki/Geonames "Geonames") to capture relationships between different geographic names and locales and associated entities. In 1998, Andrew Edmonds of Science in Finance Ltd in the UK created a system called ThinkBase that offered [fuzzy-logic](https://en.wikipedia.org/wiki/Fuzzy_logic "Fuzzy logic") based reasoning in a graphical context.[[8]](#cite_note-8) In 2007, both [DBpedia](https://en.wikipedia.org/wiki/DBpedia "DBpedia") and [Freebase](https://en.wikipedia.org/wiki/Freebase_(database) "Freebase (database)") were founded as graph-based knowledge [repositories](https://en.wikipedia.org/wiki/Repository_(version_control) "Repository (version control)") for general-purpose knowledge.[[9]](#cite_note-9) DBpedia focused exclusively on data extracted from [Wikipedia](https://en.wikipedia.org/wiki/Wikipedia "Wikipedia"), while Freebase also included a range of public datasets. Neither described themselves as a 'knowledge graph' but developed and described related concepts. In 2012, Google introduced their [Knowledge Graph](https://en.wikipedia.org/wiki/Knowledge_Graph_(Google) "Knowledge Graph (Google)"),[[10]](#cite_note-Singhal-2012-10) building on DBpedia and Freebase among other sources. They later incorporated [RDFa](https://en.wikipedia.org/wiki/RDFa "RDFa"), [Microdata](https://en.wikipedia.org/wiki/Microdata_(HTML) "Microdata (HTML)"), [JSON-LD](https://en.wikipedia.org/wiki/JSON-LD "JSON-LD") content extracted from indexed web pages, including the *[CIA World Factbook](https://en.wikipedia.org/wiki/The_World_Factbook "The World Factbook")*, [Wikidata](https://en.wikipedia.org/wiki/Wikidata "Wikidata"), and Wikipedia.[[10]](#cite_note-Singhal-2012-10)[[11]](#cite_note-11) Entity and relationship types associated with this knowledge graph have been further organized using terms from the [schema.org](https://en.wikipedia.org/wiki/Schema.org "Schema.org")[[12]](#cite_note-McCusker-12) vocabulary. The Google Knowledge Graph became a complement to string-based search within Google, and its popularity online brought the term into more common use.[[12]](#cite_note-McCusker-12) Since then, several large multinationals have advertised their use of knowledge graphs, further popularising the term. These include [Facebook](https://en.wikipedia.org/wiki/Facebook "Facebook"), [LinkedIn](https://en.wikipedia.org/wiki/LinkedIn "LinkedIn"), [Airbnb](https://en.wikipedia.org/wiki/Airbnb "Airbnb"), [Microsoft](https://en.wikipedia.org/wiki/Microsoft "Microsoft"), [Amazon](https://en.wikipedia.org/wiki/Amazon.com "Amazon.com"), [Uber](https://en.wikipedia.org/wiki/Uber "Uber") and [eBay](https://en.wikipedia.org/wiki/EBay "EBay").[[13]](#cite_note-13) In 2019, [IEEE](https://en.wikipedia.org/wiki/Institute_of_Electrical_and_Electronics_Engineers "Institute of Electrical and Electronics Engineers") combined its annual international conferences on "Big Knowledge" and "Data Mining and Intelligent Computing" into the International Conference on Knowledge Graph.[[14]](#cite_note-14) The development of large language models expanded interest in knowledge graphs as a way to structure information from unstructured text, with advances in language processing enabling their automatic or semi-automatic generation and expansion.[[15]](#cite_note-15)[[16]](#cite_note-16)[[17]](#cite_note-17) The term knowledge graph has since broadened to include the dynamically constructed and adaptive graph structures, which support retrieval, reasoning, and summarization in generative systems. Microsoft Research's [GraphRAG](https://github.com/microsoft/graphrag) (2024) exemplified this development by integrating LLM-generated graphs into retrieval-augmented generation. Definitions ----------- [[edit](/w/index.php?title=Knowledge_graph&action=edit&section=2 "Edit section: Definitions")] There is no single commonly accepted definition of a knowledge graph. Most definitions view the topic through a Semantic Web lens and include these features:[[18]](#cite_note-18) * *Flexible relations among knowledge in topical domains*: A knowledge graph (i) defines [abstract classes](https://en.wikipedia.org/wiki/Abstract_class "Abstract class") and relations of entities in a schema, (ii) mainly describes real world entities and their interrelations, organized in a graph, (iii) allows for potentially interrelating arbitrary entities with each other, and (iv) covers various topical domains.[[19]](#cite_note-19) * *General structure*: A network of entities, their semantic types, properties, and relationships.[[20]](#cite_note-20)[[21]](#cite_note-21) To represent properties, categorical or numerical values are often used. * *Supporting reasoning over inferred ontologies*: A knowledge graph acquires and integrates information into an ontology and applies a reasoner to derive new knowledge.[[3]](#cite_note-Ref1-3) There are, however, many knowledge graph representations for which some of these features are not relevant. For those knowledge graphs, this simpler definition may be more useful: * A digital structure that represents knowledge as concepts and the relationships between them (facts). A knowledge graph can include an ontology that allows both humans and machines to understand and reason about its contents.[[22]](#cite_note-22)[[23]](#cite_note-23)

{
  "evidence_unit_id": "unit:de867fff732f1cc656cc:43cc23fa52b87b4cc1d0:1:cc11a335f8242537cad2",
  "artifact_id": "artifact:de867fff732f1cc656cc:15d76103d293e31bf13c",
  "text": "For other uses, see [Knowledge graph (disambiguation)](https://en.wikipedia.org/wiki/Knowledge_graph_(disambiguation) \"Knowledge graph (disambiguation)\").\n\n[![](//upload.wikimedia.org/wikipedia/commons/thumb/5/52/Conceptual_Diagram_-_Example.svg/250px-Conceptual_Diagram_-_Example.svg.png)](https://en.wikipedia.org/wiki/File:Conceptual_Diagram_-_Example.svg)\n\nExample conceptual diagram\n\nIn [knowledge representation and reasoning](https://en.wikipedia.org/wiki/Knowledge_representation_and_reasoning \"Knowledge representation and reasoning\"), a **knowledge graph** is a [knowledge base](https://en.wikipedia.org/wiki/Knowledge_base \"Knowledge base\") that uses a [graph](https://en.wikipedia.org/wiki/Graph_(discrete_mathematics) \"Graph (discrete mathematics)\")-structured [data model](https://en.wikipedia.org/wiki/Data_model \"Data model\") or [topology](https://en.wikipedia.org/wiki/Topology \"Topology\") to represent and operate on [data](https://en.wikipedia.org/wiki/Data \"Data\"). Knowledge graphs are often used to store interlinked descriptions of [entities](https://en.wikipedia.org/wiki/Named_entity \"Named entity\") –  objects, events, situations or abstract concepts –  while also encoding the free-form [semantics](https://en.wikipedia.org/wiki/Semantics \"Semantics\") or relationships underlying these entities.[[1]](#cite_note-1)[[2]](#cite_note-2)\n\nSince the development of the [Semantic Web](https://en.wikipedia.org/wiki/Semantic_Web \"Semantic Web\"), knowledge graphs have often been associated with [linked open data](https://en.wikipedia.org/wiki/Linked_data \"Linked data\") projects, focusing on the connections between [concepts](https://en.wikipedia.org/wiki/Concept \"Concept\") and entities.[[3]](#cite_note-Ref1-3)[[4]](#cite_note-4) They are also historically associated with and used by [search engines](https://en.wikipedia.org/wiki/Search_engine \"Search engine\") such as [Google](https://en.wikipedia.org/wiki/Knowledge_Graph_(Google) \"Knowledge Graph (Google)\"), [Bing](https://en.wikipedia.org/wiki/Bing_(search_engine) \"Bing (search engine)\"), and [Yahoo](https://en.wikipedia.org/wiki/Yahoo \"Yahoo\"); [knowledge engines](https://en.wikipedia.org/wiki/Knowledge_engine \"Knowledge engine\") and question-answering services such as [WolframAlpha](https://en.wikipedia.org/wiki/WolframAlpha \"WolframAlpha\"), Apple's [Siri](https://en.wikipedia.org/wiki/Siri \"Siri\"), and [Amazon Alexa](https://en.wikipedia.org/wiki/Amazon_Alexa \"Amazon Alexa\"); and [social networks](https://en.wikipedia.org/wiki/Social_network \"Social network\") such as [LinkedIn](https://en.wikipedia.org/wiki/LinkedIn \"LinkedIn\") and [Facebook](https://en.wikipedia.org/wiki/Facebook \"Facebook\").\n\nRecent developments in data science and [machine learning](https://en.wikipedia.org/wiki/Machine_learning \"Machine learning\"), particularly in [graph neural networks](https://en.wikipedia.org/wiki/Graph_neural_network \"Graph neural network\"), representation learning, and machine learning, have broadened the scope of knowledge graphs beyond their traditional use in search engines and [recommender systems](https://en.wikipedia.org/wiki/Recommender_system \"Recommender system\"). They are increasingly used in scientific research, with notable applications in fields such as [genomics](https://en.wikipedia.org/wiki/Genomics \"Genomics\"), [proteomics](https://en.wikipedia.org/wiki/Proteomics \"Proteomics\"), and [systems biology](https://en.wikipedia.org/wiki/Systems_biology \"Systems biology\").[[5]](#cite_note-5)\n\nHistory\n-------\n\n[[edit](/w/index.php?title=Knowledge_graph&action=edit&section=1 \"Edit section: History\")]\n\nThe term was coined as early as 1972 by the Austrian [linguist](https://en.wikipedia.org/wiki/Linguistics \"Linguistics\") [Edgar W. Schneider](https://en.wikipedia.org/wiki/Edgar_W._Schneider \"Edgar W. Schneider\"), in a discussion of how to build modular instructional systems for courses.[[6]](#cite_note-6) In the late 1980s, the [University of Groningen](https://en.wikipedia.org/wiki/University_of_Groningen \"University of Groningen\") and [University of Twente](https://en.wikipedia.org/wiki/University_of_Twente \"University of Twente\") jointly began a project called Knowledge Graphs, focusing on the design of [semantic networks](https://en.wikipedia.org/wiki/Semantic_network \"Semantic network\") with edges restricted to a limited set of relations, to facilitate [algebras on the graph](https://en.wikipedia.org/wiki/Graph_algebra \"Graph algebra\").[[7]](#cite_note-7) In subsequent decades, the distinction between semantic networks and knowledge graphs was blurred.\n\nSome early knowledge graphs were topic-specific. In 1985, [Wordnet](https://en.wikipedia.org/wiki/Wordnet \"Wordnet\") was founded, capturing semantic relationships between words and meanings – an application of this idea to language itself. In 2005, Marc Wirk founded [Geonames](https://en.wikipedia.org/wiki/Geonames \"Geonames\") to capture relationships between different geographic names and locales and associated entities. In 1998, Andrew Edmonds of Science in Finance Ltd in the UK created a system called ThinkBase that offered [fuzzy-logic](https://en.wikipedia.org/wiki/Fuzzy_logic \"Fuzzy logic\") based reasoning in a graphical context.[[8]](#cite_note-8)\n\nIn 2007, both [DBpedia](https://en.wikipedia.org/wiki/DBpedia \"DBpedia\") and [Freebase](https://en.wikipedia.org/wiki/Freebase_(database) \"Freebase (database)\") were founded as graph-based knowledge [repositories](https://en.wikipedia.org/wiki/Repository_(version_control) \"Repository (version control)\") for general-purpose knowledge.[[9]](#cite_note-9) DBpedia focused exclusively on data extracted from [Wikipedia](https://en.wikipedia.org/wiki/Wikipedia \"Wikipedia\"), while Freebase also included a range of public datasets. Neither described themselves as a 'knowledge graph' but developed and described related concepts.\n\nIn 2012, Google introduced their [Knowledge Graph](https://en.wikipedia.org/wiki/Knowledge_Graph_(Google) \"Knowledge Graph (Google)\"),[[10]](#cite_note-Singhal-2012-10) building on DBpedia and Freebase among other sources. They later incorporated [RDFa](https://en.wikipedia.org/wiki/RDFa \"RDFa\"), [Microdata](https://en.wikipedia.org/wiki/Microdata_(HTML) \"Microdata (HTML)\"), [JSON-LD](https://en.wikipedia.org/wiki/JSON-LD \"JSON-LD\") content extracted from indexed web pages, including the *[CIA World Factbook](https://en.wikipedia.org/wiki/The_World_Factbook \"The World Factbook\")*, [Wikidata](https://en.wikipedia.org/wiki/Wikidata \"Wikidata\"), and Wikipedia.[[10]](#cite_note-Singhal-2012-10)[[11]](#cite_note-11) Entity and relationship types associated with this knowledge graph have been further organized using terms from the [schema.org](https://en.wikipedia.org/wiki/Schema.org \"Schema.org\")[[12]](#cite_note-McCusker-12) vocabulary. The Google Knowledge Graph became a complement to string-based search within Google, and its popularity online brought the term into more common use.[[12]](#cite_note-McCusker-12)\n\nSince then, several large multinationals have advertised their use of knowledge graphs, further popularising the term. These include [Facebook](https://en.wikipedia.org/wiki/Facebook \"Facebook\"), [LinkedIn](https://en.wikipedia.org/wiki/LinkedIn \"LinkedIn\"), [Airbnb](https://en.wikipedia.org/wiki/Airbnb \"Airbnb\"), [Microsoft](https://en.wikipedia.org/wiki/Microsoft \"Microsoft\"), [Amazon](https://en.wikipedia.org/wiki/Amazon.com \"Amazon.com\"), [Uber](https://en.wikipedia.org/wiki/Uber \"Uber\") and [eBay](https://en.wikipedia.org/wiki/EBay \"EBay\").[[13]](#cite_note-13)\n\nIn 2019, [IEEE](https://en.wikipedia.org/wiki/Institute_of_Electrical_and_Electronics_Engineers \"Institute of Electrical and Electronics Engineers\") combined its annual international conferences on \"Big Knowledge\" and \"Data Mining and Intelligent Computing\" into the International Conference on Knowledge Graph.[[14]](#cite_note-14)\n\nThe development of large language models expanded interest in knowledge graphs as a way to structure information from unstructured text, with advances in language processing enabling their automatic or semi-automatic generation and expansion.[[15]](#cite_note-15)[[16]](#cite_note-16)[[17]](#cite_note-17) The term knowledge graph has since broadened to include the dynamically constructed and adaptive graph structures, which support retrieval, reasoning, and summarization in generative systems. Microsoft Research's [GraphRAG](https://github.com/microsoft/graphrag) (2024) exemplified this development by integrating LLM-generated graphs into retrieval-augmented generation.\n\nDefinitions\n-----------\n\n[[edit](/w/index.php?title=Knowledge_graph&action=edit&section=2 \"Edit section: Definitions\")]\n\nThere is no single commonly accepted definition of a knowledge graph. Most definitions view the topic through a Semantic Web lens and include these features:[[18]](#cite_note-18)\n\n* *Flexible relations among knowledge in topical domains*: A knowledge graph (i) defines [abstract classes](https://en.wikipedia.org/wiki/Abstract_class \"Abstract class\") and relations of entities in a schema, (ii) mainly describes real world entities and their interrelations, organized in a graph, (iii) allows for potentially interrelating arbitrary entities with each other, and (iv) covers various topical domains.[[19]](#cite_note-19)\n* *General structure*: A network of entities, their semantic types, properties, and relationships.[[20]](#cite_note-20)[[21]](#cite_note-21) To represent properties, categorical or numerical values are often used.\n* *Supporting reasoning over inferred ontologies*: A knowledge graph acquires and integrates information into an ontology and applies a reasoner to derive new knowledge.[[3]](#cite_note-Ref1-3)\n\nThere are, however, many knowledge graph representations for which some of these features are not relevant. For those knowledge graphs, this simpler definition may be more useful:\n\n* A digital structure that represents knowledge as concepts and the relationships between them (facts). A knowledge graph can include an ontology that allows both humans and machines to understand and reason about its contents.[[22]](#cite_note-22)[[23]](#cite_note-23)",
  "access_class": "public",
  "heading": "Document",
  "observed_at": "2026-07-18T21:40:59.214Z",
  "state": "active"
}