Ontology learning

<p>Ontology learning (ontology extraction,ontology augmentation generation, ontology generation, or ontology acquisition) is the automatic or semi-automatic creation of <a href="/facts/Ontology_(information_science)/SQlhl6lp">ontologies</a>, including extracting the corresponding <a href="/facts/Domain_of_discourse/T1WRQxx6">domain's</a> terms and the relationships between the <a href="/facts/Conceptualization_(information_science)/jCUJBHWI">concepts</a> that these terms represent from a <a href="/facts/Text_corpus/bYeEQQdh">corpus</a> of natural language text, and encoding them with an <a href="/facts/Ontology_language/yXLTb2z3">ontology language</a> for easy retrieval. As <a href="/facts/Ontology_engineering/XzBfzVuC">building ontologies</a> manually is extremely labor-intensive and time-consuming, there is great motivation to automate the process.
</p><p>Typically, the process starts by <a href="/facts/Terminology_extraction/TRjKaWct">extracting terms</a> and concepts or <a href="/facts/Noun_phrase/mmz3tJt9">noun phrases</a> from plain text using linguistic processors such as <a href="/facts/Part-of-speech_tagging/BYAPmH3r">part-of-speech tagging</a> and <a href="/facts/Phrase_chunking/QXqsV678">phrase chunking</a>. Then statistical 
or symbolic
techniques are used to extract <a href="/facts/First-order_logic/lcISqWIg">relation signatures</a>, often based on pattern-based or definition-based hypernym extraction techniques.
</p>

Ontology learning open-in-new

Ontology learning