knowledge graph
In plain wordsA huge web of facts stored as connections between things — 'Paris is the capital of France', 'Marie Curie won a Nobel Prize' — that computers can search, combine and reason over.
A knowledge baseknowledge base, n.: A store of facts and rules that a program can reason with: like a database, but organized so that new conclusions can be worked out from it, not just looked up. that stores each fact as a labeled, directed link between two entities — a subject–predicate–object triple such as (Marie Curie, award received, Nobel Prize in Physics) — so that the entities (people, places, products, concepts) and their relations together form one large network. Entities carry types and attributes, and an ontologyontology, n.: A carefully defined vocabulary for a subject such as medicine, spelling out each kind of thing, how kinds nest inside one another, and which connections between them are allowed. or schema usually constrains which relations are allowed. Knowledge graphs are stored in RDFResource Description Framework, prop. n.: A web standard for writing facts as simple three-part statements — subject, relation, object — using web addresses as names, so that data from different sources can be merged automatically. triple stores or property-graph databases and queried with languages such as SPARQL or Cypher. They power search-engine answer panels, recommendation, fraud detection, drug discovery and enterprise data integration. Building them involves information extractioninformation extraction, n.: Automatically pulling structured facts out of ordinary text — who did what, when, where and for how much — and putting them into a table or database., entity resolution and linking; missing facts can be predicted with embeddingsknowledge graph embedding, n.: A way of turning the things and relations in a knowledge graph into lists of numbers, so a computer can guess missing facts from patterns, such as predicting where someone was born. or learned rules. In language-modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. systems they supply structured, checkable facts to GraphRAGGraphRAG, prop. n.: A way of letting AI answer questions about a big pile of documents by first turning them into a map of people, things and their connections, then using that map to find and summarize answers. and other retrieval-augmentedretrieval-augmented generation, n.: Giving an AI an open-book exam: before answering, it looks up relevant passages in a set of documents and bases its reply on them, often citing where each point came from. pipelines.
We ground the chatbot's answers in the product knowledge graph, so every claim links back to a node.
Origin
The phrase appeared sporadically from the 1970s, including a Dutch research program on knowledge graphs in the 1980s, but became widespread after Google announced its Knowledge Graph in May 2012 under the slogan 'things, not strings'. Earlier open graphs such as DBpedia and Freebase (both 2007) fed the trend, and WikidataWikidata, prop. n.: A free, shared collection of facts about millions of things — people, places, books, chemicals — kept by volunteers in the Wikipedia family and laid out so programs can use it as easily as people. launched later in 2012.
Usage note
Usage is loose: some reserve the term for graphs with a formal ontology and reasoning, while industry applies it to almost any graph of entities. Compare semantic networksemantic network, n.: A way of storing knowledge as a web of concepts joined by labeled arrows, such as 'canary — is a — bird' and 'bird — can — fly', so a program can follow the links., its older AI ancestor, and ontologyontology, n.: A carefully defined vocabulary for a subject such as medicine, spelling out each kind of thing, how kinds nest inside one another, and which connections between them are allowed., which supplies the schema rather than the instance data.
Related
- knowledge baseknowledge base, n.: A store of facts and rules that a program can reason with: like a database, but organized so that new conclusions can be worked out from it, not just looked up.
- ontologyontology, n.: A carefully defined vocabulary for a subject such as medicine, spelling out each kind of thing, how kinds nest inside one another, and which connections between them are allowed.
- semantic websemantic web, n.: An effort to label data on the web in a standard, precise way so that programs, not only human readers, can tell what it means and connect facts drawn from many different sites.
- Resource Description FrameworkResource Description Framework, prop. n.: A web standard for writing facts as simple three-part statements — subject, relation, object — using web addresses as names, so that data from different sources can be merged automatically.
- knowledge graph embeddingknowledge graph embedding, n.: A way of turning the things and relations in a knowledge graph into lists of numbers, so a computer can guess missing facts from patterns, such as predicting where someone was born.
- GraphRAGGraphRAG, prop. n.: A way of letting AI answer questions about a big pile of documents by first turning them into a map of people, things and their connections, then using that map to find and summarize answers.
- WikidataWikidata, prop. n.: A free, shared collection of facts about millions of things — people, places, books, chemicals — kept by volunteers in the Wikipedia family and laid out so programs can use it as easily as people.
- semantic networksemantic network, n.: A way of storing knowledge as a web of concepts joined by labeled arrows, such as 'canary — is a — bird' and 'bird — can — fly', so a program can follow the links.
Sources
- Singhal, A. (2012). Introducing the Knowledge Graph: Things, Not Strings. Google Official Blog, 16 May 2012.
- Hogan, A. et al. (2021). Knowledge Graphs. ACM Computing Surveys 54(4), 1–37.
Mentioned in
agent memoryagent memory, n.: The ways an AI assistant or agent keeps information — your preferences, past conversations, lessons from earlier tasks — and brings it back later, beyond what fits in a single conversation., symbolic AIsymbolic AI, n.: An approach to AI that writes knowledge down as human-readable symbols and rules, such as 'every bird has wings', and reasons by manipulating them step by step instead of learning patterns from data., graph neural networkgraph neural network, n.: A neural network for data shaped like a web of connections, such as molecules, social networks or road maps, in which each point repeatedly gathers information from its neighbors., GraphRAGGraphRAG, prop. n.: A way of letting AI answer questions about a big pile of documents by first turning them into a map of people, things and their connections, then using that map to find and summarize answers., information extractioninformation extraction, n.: Automatically pulling structured facts out of ordinary text — who did what, when, where and for how much — and putting them into a table or database., knowledge baseknowledge base, n.: A store of facts and rules that a program can reason with: like a database, but organized so that new conclusions can be worked out from it, not just looked up., knowledge representationknowledge representation, n.: The study of how to record knowledge about objects, categories, rules and events in a precise format, so that a program can reason with it and reach conclusions nobody typed in., long-term memorylong-term memory, n.: Memory that lasts. For an AI agent, it means information saved outside the conversation — facts, past events, learned procedures — that can be looked up and used again days or months later., named-entity recognitionnamed-entity recognition, n.: Software that spots names and similar expressions in text and labels each by type: in 'Maria flew from Lagos to Paris on Monday', it marks a person, two cities and a date., neuro-symbolic AIneuro-symbolic AI, n.: AI that combines neural networks, which learn patterns from data, with symbolic methods, which follow explicit rules and logic, aiming for systems that both learn flexibly and reason reliably., ontologyontology, n.: A carefully defined vocabulary for a subject such as medicine, spelling out each kind of thing, how kinds nest inside one another, and which connections between them are allowed., semantic networksemantic network, n.: A way of storing knowledge as a web of concepts joined by labeled arrows, such as 'canary — is a — bird' and 'bird — can — fly', so a program can follow the links., semantic searchsemantic search, n.: Search that understands what you mean, not just the words you typed, so a search for 'cheap places to stay' also finds 'budget hotels'., semantic websemantic web, n.: An effort to label data on the web in a standard, precise way so that programs, not only human readers, can tell what it means and connect facts drawn from many different sites., WikidataWikidata, prop. n.: A free, shared collection of facts about millions of things — people, places, books, chemicals — kept by volunteers in the Wikipedia family and laid out so programs can use it as easily as people., WordNetWordNet, prop. n.: A hand-made map of English vocabulary in which words sharing a meaning are bundled together and the bundles are linked — recording, say, that a dog is a kind of animal and a wheel is part of a car., ConceptNetConceptNet, prop. n.: A free, crowd-built network of everyday facts — 'a knife is used for cutting', 'rain makes things wet' — in many languages, which AI systems use to make sense of ordinary situations., DatalogDatalog, prop. n.: A small logic language for asking questions of a database with rules, including rules that refer to themselves, such as 'X is an ancestor of Y if X is a parent of an ancestor of Y'., knowledge engineeringknowledge engineering, n.: The craft of collecting what human experts know — through interviews, observation and documents — and turning it into rules or structured data that a computer system can use., knowledge graph embeddingknowledge graph embedding, n.: A way of turning the things and relations in a knowledge graph into lists of numbers, so a computer can guess missing facts from patterns, such as predicting where someone was born., Resource Description FrameworkResource Description Framework, prop. n.: A web standard for writing facts as simple three-part statements — subject, relation, object — using web addresses as names, so that data from different sources can be merged automatically., Web Ontology LanguageWeb Ontology Language, prop. n.: A web standard language for writing ontologies — precise definitions of categories and their relationships — so that software can check them for contradictions and infer new facts.