The experience from the Textual Entailment con- tests has revealed that an important part of entail- ment relationships requires extensive world kno w- ledge, which is unlikely to be found in a repository, because this information is bound to specific co n- texts. In this paper we present an unsupervised and language independent method of acquiring quasi- similar pairs which capture this particular type of knowledge.
Learning Quasi-Similar Pairs for Textual Entailment
Popescu, Octavian;Pianta, Emanuele;Magnini, Bernardo
2011-01-01
Abstract
The experience from the Textual Entailment con- tests has revealed that an important part of entail- ment relationships requires extensive world kno w- ledge, which is unlikely to be found in a repository, because this information is bound to specific co n- texts. In this paper we present an unsupervised and language independent method of acquiring quasi- similar pairs which capture this particular type of knowledge.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
