QUICK FACTS
Created Jan 0001
Status Verified Sarcastic
Type Existential Dread
Keywords
multilingual, knowledge graph, ontology, dictionary, sapienza university of rome, roberto navigli, wikipedia, lexicon, english language

BabelNet

“is a multilingual lexical‑semantic knowledge graph, ontology) and encyclopedic dictionary that was developed at the NLP group of the Sapienza University of...”

Contents
  • 1. Overview
  • 2. Etymology
  • 3. Cultural Impact

BabelNet

BabelNet is a multilingual lexical‑semantic knowledge graph , ontology and encyclopedic dictionary that was developed at the NLP group of the Sapienza University of Rome under the supervision of Roberto Navigli . [1] [2] It was automatically created by linking Wikipedia to the most popular computational lexicon of the English language , WordNet . The integration was performed using an automatic mapping and by filling in lexical gaps in resource‑poor languages through the use of statistical machine translation . The result is an encyclopedic dictionary that provides concepts and named entities lexicalized in many languages and interconnects them with a large number of semantic relations . Additional lexicalizations and definitions are contributed by linking to freely licensed wordnets, OmegaWiki , the English Wiktionary , Wikidata , FrameNet , VerbNet and other resources. In analogy to WordNet , BabelNet groups words in different languages into sets of synonyms called Babel synsets . For each Babel synset the system supplies short glosses in numerous languages that are harvested from both WordNet and Wikipedia .

Statistics of BabelNet

As of December 2023 [update] , BabelNet (version 5.3) covers 600 languages . It contains almost 23 million synsets and around 1.7 billion word senses (regardless of their language). Each Babel synset contains, on average, two synonyms per language, i.e., two word senses . The semantic network includes all lexico‑semantic relations from WordNet (hypernymy and hyponymy , meronymy and holonymy , antonymy and synonymy , etc.), amounting to roughly 364 000 relation edges, as well as an underspecified relatedness relation extracted from Wikipedia (approximately 1.9 billion edges). [1] Version 5.3 also links around 61 million images to Babel synsets and provides a Lemon RDF encoding of the resource, [3] which is accessible through a SPARQL endpoint . Additionally, 2.67 million synsets are assigned domain labels .

Applications

BabelNet has been demonstrated to empower a variety of multilingual natural language processing tasks. Its richly lexicalized knowledge yields state‑of‑the‑art performance in several areas, including:

Prizes and acknowledgments

The resource has received several distinctions, among them the META prize 2015 for “groundbreaking work in overcoming language barriers through a multilingual lexicalised semantic network and ontology making use of heterogeneous data sources”. The Artificial Intelligence Journal article that introduced BabelNet [1] was awarded the Prominent Paper Award in 2017. [9] Moreover, BabelNet was highlighted in a Time magazine feature [10] that discussed innovative lexical knowledge resources on the Web.

See also