Wordnet

WordNet is a large lexical database of English designed to support natural language processing and linguistic research. It groups English words into sets of cognitive synonyms called synsets, where each synset represents a single concept. Each synset has a short definition (gloss) and one or more usage examples. Words are linked to their synsets by part of speech, and the synsets themselves are connected through various semantic and lexical relations, such as hypernymy and hyponymy (is-a relationships), meronymy and holonymy (part-whole relationships), as well as antonymy, entailment, and more.

Key features include the organization of words by senses rather than by form, the explicit representation of relationships between senses, and the presence of cross-sense definitions and usage examples. This structure enables tasks such as measuring semantic similarity, performing word sense disambiguation, and supporting information retrieval and machine translation.

WordNet was developed at Princeton University in the Cognitive Science and Psychology communities, with principal contributions from George A. Miller and Christiane Fellbaum. The project began in the 1980s, and the first public release appeared in 1990, followed by several major updates. Since then, WordNet has become a foundational resource in natural language processing and linguistics, widely used in academic research and commercial applications.

In addition to the original English WordNet, related efforts have produced multilingual and multilingual-linked resources (for example, EuroWordNet and Open Multilingual WordNet) built on similar concepts. WordNet is often accessed programmatically via standard NLP tooling and libraries, such as those in Python, which provide interfaces to the database.