[[edit](/w/index.php?title=Knowledge_graph&action=edit&section=3 "Edit section: Implementations")] In addition to the above examples, the term has been used to describe open knowledge projects such as [YAGO](https://en.wikipedia.org/wiki/YAGO_(database) "YAGO (database)") and Wikidata; federations like the Linked Open Data cloud;[[24]](#cite_note-24) a range of commercial search tools, including Yahoo's semantic search assistant Spark, Google's [Knowledge Graph](https://en.wikipedia.org/wiki/Knowledge_Graph_(Google) "Knowledge Graph (Google)"), and Microsoft's Satori; and the LinkedIn and Facebook entity graphs.[[3]](#cite_note-Ref1-3) The term is also used in the context of [note-taking software](https://en.wikipedia.org/wiki/Note-taking_software "Note-taking software") applications that allow a user to build a [personal knowledge graph](https://en.wikipedia.org/wiki/Personal_knowledge_graph "Personal knowledge graph").[[25]](#cite_note-25) The popularization of knowledge graphs and their accompanying methods have led to the development of graph databases such as Neo4j,[[26]](#cite_note-26) GraphDB[[27]](#cite_note-27) and [AgensGraph](https://en.wikipedia.org/wiki/AgensGraph?action=edit&redlink=1 "AgensGraph (page does not exist)").[[28]](#cite_note-28) These graph databases allow users to easily store data as entities and their interrelationships, and facilitate operations such as data reasoning, node embedding, and ontology development on knowledge bases. In contrast, virtual knowledge graphs do not store information in specialized databases.[[29]](#cite_note-29) They rely on an underlying relational database or data lake to answer queries on the graph. Such a virtual knowledge graph system must be properly configured in order to answer the queries correctly. This specific configuration is done through a set of mappings that define the relationship between the elements of the data source and the structure and ontology of the virtual knowledge graph.[[30]](#cite_note-30) Using a knowledge graph for reasoning over data ----------------------------------------------- [[edit](/w/index.php?title=Knowledge_graph&action=edit&section=4 "Edit section: Using a knowledge graph for reasoning over data")] Main article: [Ontology (information science)](https://en.wikipedia.org/wiki/Ontology_(information_science) "Ontology (information science)") A knowledge graph formally represents semantics by describing entities and their relationships.[[31]](#cite_note-31) Knowledge graphs may make use of [ontologies](https://en.wikipedia.org/wiki/Ontology_(information_science) "Ontology (information science)") as a schema layer. By doing this, they allow [logical inference](https://en.wikipedia.org/wiki/Inference "Inference") for retrieving [implicit knowledge](https://en.wikipedia.org/wiki/Implicit_knowledge "Implicit knowledge") rather than only allowing queries requesting explicit knowledge.[[32]](#cite_note-32) In order to allow the use of knowledge graphs in various machine learning tasks, several methods for deriving latent feature representations of entities and relations have been devised.[[33]](#cite_note-33) These knowledge graph embeddings allow them to be connected to machine learning methods that require feature vectors like [word embeddings](https://en.wikipedia.org/wiki/Word_embedding "Word embedding"). This can complement other estimates of conceptual similarity.[[34]](#cite_note-34)[[35]](#cite_note-35) Models for generating useful knowledge graph embeddings are commonly the domain of graph neural networks (GNNs).[[36]](#cite_note-36) GNNs are deep learning architectures that comprise edges and nodes, which correspond well to the entities and relationships of knowledge graphs. The topology and data structures afforded by GNNs provide a convenient domain for semi-supervised learning, wherein the network is trained to predict the value of a node embedding (provided a group of adjacent nodes and their edges) or edge (provided a pair of nodes). These tasks serve as fundamental abstractions for more complex tasks such as knowledge graph reasoning and alignment.[[37]](#cite_note-37)

{
  "evidence_unit_id": "unit:de867fff732f1cc656cc:663b5d3875dec98af0ca:1:ecef767b65758a81f2f9",
  "artifact_id": "artifact:de867fff732f1cc656cc:15d76103d293e31bf13c",
  "text": "[[edit](/w/index.php?title=Knowledge_graph&action=edit&section=3 \"Edit section: Implementations\")]\n\nIn addition to the above examples, the term has been used to describe open knowledge projects such as [YAGO](https://en.wikipedia.org/wiki/YAGO_(database) \"YAGO (database)\") and Wikidata; federations like the Linked Open Data cloud;[[24]](#cite_note-24) a range of commercial search tools, including Yahoo's semantic search assistant Spark, Google's [Knowledge Graph](https://en.wikipedia.org/wiki/Knowledge_Graph_(Google) \"Knowledge Graph (Google)\"), and Microsoft's Satori; and the LinkedIn and Facebook entity graphs.[[3]](#cite_note-Ref1-3)\n\nThe term is also used in the context of [note-taking software](https://en.wikipedia.org/wiki/Note-taking_software \"Note-taking software\") applications that allow a user to build a [personal knowledge graph](https://en.wikipedia.org/wiki/Personal_knowledge_graph \"Personal knowledge graph\").[[25]](#cite_note-25)\n\nThe popularization of knowledge graphs and their accompanying methods have led to the development of graph databases such as Neo4j,[[26]](#cite_note-26) GraphDB[[27]](#cite_note-27) and [AgensGraph](https://en.wikipedia.org/wiki/AgensGraph?action=edit&redlink=1 \"AgensGraph (page does not exist)\").[[28]](#cite_note-28) These graph databases allow users to easily store data as entities and their interrelationships, and facilitate operations such as data reasoning, node embedding, and ontology development on knowledge bases.\n\nIn contrast, virtual knowledge graphs do not store information in specialized databases.[[29]](#cite_note-29) They rely on an underlying relational database or data lake to answer queries on the graph. Such a virtual knowledge graph system must be properly configured in order to answer the queries correctly. This specific configuration is done through a set of mappings that define the relationship between the elements of the data source and the structure and ontology of the virtual knowledge graph.[[30]](#cite_note-30)\n\nUsing a knowledge graph for reasoning over data\n-----------------------------------------------\n\n[[edit](/w/index.php?title=Knowledge_graph&action=edit&section=4 \"Edit section: Using a knowledge graph for reasoning over data\")]\n\nMain article: [Ontology (information science)](https://en.wikipedia.org/wiki/Ontology_(information_science) \"Ontology (information science)\")\n\nA knowledge graph formally represents semantics by describing entities and their relationships.[[31]](#cite_note-31) Knowledge graphs may make use of [ontologies](https://en.wikipedia.org/wiki/Ontology_(information_science) \"Ontology (information science)\") as a schema layer. By doing this, they allow [logical inference](https://en.wikipedia.org/wiki/Inference \"Inference\") for retrieving [implicit knowledge](https://en.wikipedia.org/wiki/Implicit_knowledge \"Implicit knowledge\") rather than only allowing queries requesting explicit knowledge.[[32]](#cite_note-32)\n\nIn order to allow the use of knowledge graphs in various machine learning tasks, several methods for deriving latent feature representations of entities and relations have been devised.[[33]](#cite_note-33) These knowledge graph embeddings allow them to be connected to machine learning methods that require feature vectors like [word embeddings](https://en.wikipedia.org/wiki/Word_embedding \"Word embedding\"). This can complement other estimates of conceptual similarity.[[34]](#cite_note-34)[[35]](#cite_note-35)\n\nModels for generating useful knowledge graph embeddings are commonly the domain of graph neural networks (GNNs).[[36]](#cite_note-36) GNNs are deep learning architectures that comprise edges and nodes, which correspond well to the entities and relationships of knowledge graphs. The topology and data structures afforded by GNNs provide a convenient domain for semi-supervised learning, wherein the network is trained to predict the value of a node embedding (provided a group of adjacent nodes and their edges) or edge (provided a pair of nodes). These tasks serve as fundamental abstractions for more complex tasks such as knowledge graph reasoning and alignment.[[37]](#cite_note-37)",
  "access_class": "public",
  "heading": "Implementations",
  "observed_at": "2026-07-18T21:40:59.214Z",
  "state": "active"
}