Integración del reconocimiento de la implicación textual en tareas automáticas de resúmenes de textos
ISSN: 1135-5948
Année de publication: 2008
Número: 41
Pages: 183-190
Type: Article
D'autres publications dans: Procesamiento del lenguaje natural
Résumé
This paper presents a preliminary study of the usefulness of applying textual entailment recognition in text summarization tasks. Two different approaches are proposed. The first one relies on the word's frequency in order to select the most relevant sentences to include in the final summary, whereas the second one combines textual entailment recognition together with the aforementioned approach so that we can reduce the size of the original text or texts first. Results show that this combination has been appropriate, obtaining an increase of 5.85 % for single-document and of 6.80 % for multi-document when textual entailment and the frequency of words are applied together.