Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Distributional–relational database
open-in-new
<p>A distributional–relational database, or word-vector database, is a <a href="/facts/Database_Management_System/VYcpkevb">database management system</a> (DBMS) that uses distributional <a href="/facts/Word_embedding/7uRcBPqo">word-vector</a> representations to enrich the semantics of <a href="/facts/Data_model/j1YyzpJ3">structured data</a>. </p><p>As distributional word-vectors can be built automatically from large-scale <a href="/facts/Text_corpus/bYeEQQdh">corpora</a>, this enrichment supports the construction of databases which can embed large-scale commonsense background knowledge into their operations. Distributional-Relational models can be applied to the construction of <a href="/facts/Schema-agnostic_databases/nbMELyjH">schema-agnostic databases</a> (databases in which users can query the data without being aware of its <a href="/facts/Database_schema/DBq8fooY">schema</a>), <a href="/facts/Semantic_search/f7lwjtwf">semantic search</a>, schema-integration and <a href="/facts/Inductive_reasoning/EJ992JJ9">inductive</a> and <a href="/facts/Abductive_reasoning/QoHNqvKT">abductive reasoning</a> as well as different applications in which a semantically flexible knowledge representation model is needed. The main advantage of distributional–relational models over purely logical / <a href="/facts/Semantic_Web/NFHo9dB7">semantic web</a> models is the fact that the core semantic associations can be automatically captured from corpora, in contrast to the definition of manually curated <a href="/facts/Ontology/QJKTsEIJ">ontologies</a> and rule knowledge bases. </p>