Rol semantikoen etiketatze automatikoa

• Autores: Haritz Salaberri, Olatz Arregi Uriarte, Beñat Zapirain Sierra
• Localización: Ekaia: Euskal Herriko Unibertsitateko zientzi eta teknologi aldizkaria, ISSN 0214-9001, Nº. 27, 2014, págs. 297-313
• Idioma: euskera
• Enlaces
  • Texto completo
• Resumen
  • English

    The main task of semantic role labeling (SRL), sometimes also called shallow semantic parsing , is to detect the semantic relations hold among the predicate of a sentence and its associated participants and properties and the classification into their spec ific roles . Perforrrung sentence-level semantic analysis can help determine who did whar to whom, where, when, and how within an event . The predicate of a clause (typically a verb) establishes what took place , and other sentence constituents express the participants in the event (such as who and where), as well as further event properties (such as when and how). The information provided by semantic roles is crucial in order to process texts automatically, and in addition to the applications in Natural Language Processing (NLP), semantic roles can help improve Internet search engines, question answering and translation systems. Nowadays , roles are on the edge regarding information extraction and social network research tasks.

  • euskara

    Rol semantikoen etiketatze automatikoa (SRL), azaleko anali si semantikoa ere deitua , hi zkuntzalaritza konputazionalaren ikerlerro garrantzitsua da eta bertan, zehatz finkatu nahj dira testu bateko gertakarietan, ekjntza eta honetan parte hartzen dutenen arteko erlazio semantikoak edo rolak; berez, nork, nori, zer egin zion, non eta noiz gertatu den jakin nahi da. Rolek eskaintzen duten informazioak berebiziko garrantzia dauka testuak automatikoki prozesatu eta ulertzeko bidean. Ataza hau zeresan handia ematen ari da hizkuntzaren prozesamenduan ez ezik, besteak beste, Interneteko bilatzaileetan , itzulpen automatikoko eta galdera-erantzun sistemetan, sare sozialen azterketa automatikoan, eta dokumentuen informazio erauzketan.