Hypernymy and hyponymy
Hypernymy and hyponymy are the semantic relations between a generic term (hypernym) and a specific instance of it (hyponym). The hypernym is also called a supertype, umbrella term, or blanket term.[1][2][3][4] The hyponym is a subtype of the hypernym. The semantic field of the hyponym is included within that of the hypernym.[5] For example, pigeon, crow, and hen are all hyponyms of bird and animal; bird and animal are both hypernyms of pigeon, crow, and hen.[6]
Etymology[edit]
Hyperonym and hypernym mean the same thing, with both in use by linguists. The form hypernym interprets the -o- of hyponym as a part of hypo, such as in hypertension and hypotension. However, etymologically the -o- is part of the Greek stem ónoma. In other combinations with this stem, e.g. synonym, it is never elided. Therefore, hyperonym is etymologically more faithful than hypernym.[15] Hyperonymy is used, for instance, by John Lyons, who does not mention hypernymy and prefers superordination.[16] The nominalization hyperonymy is rarely used, because the neutral term to refer to the relationship is hyponymy.
Usage[edit]
Computer science often terms this relationship an "is-a" relationship. For example, the phrase "Red is-a color" can be used to describe the hyponymic relationship between red and color.
Hyponymy is the most frequently encoded relation among synsets used in lexical databases such as WordNet. These semantic relations can also be used to compare semantic similarity by judging the distance between two synsets and to analyse anaphora.
As a hypernym can be understood as a more general word than its hyponym, the relation is used in semantic compression by generalization to reduce a level of specialization.
The notion of hyponymy is particularly relevant to language translation, as hyponyms are very common across languages. For example, in Japanese the word for older brother is ani (兄), and the word for younger brother is otōto (弟). An English-to-Japanese translator presented with a phrase containing the English word brother would have to choose which Japanese word equivalent to use. This would be difficult, because abstract information (such as the speakers' relative ages) is often not available during machine translation.