In this paper we will present a system for Question Answering called GETARUNS, in its deep version applicable to closed domains, that is to say domains for which the lexical semantics is fully specified and does not have to be induced. In addition, no ontology is needed: semantic re- lations are derived from linguistic relations encoded in the syntax. The main tenet of the system is that it is possible to produce consistent seman- tic representations using a strict linguistic approach without resorting to extralinguistic knowledge sources. The paper will briefly present the low level component which is responsible for pronominal binding, quantifier raising and temporal interpretation. Then it will discuss in more detail the high level component where a Discourse Model is created from text. The system has been evaluated on a wide variety of texts from closed domains, producing full and accurate parsing, semantics and anaphora resolution for all sentences.
Answering Why-Questions in Closed Domains from a Discourse Model
DELMONTE, Rodolfo;
2008-01-01
Abstract
In this paper we will present a system for Question Answering called GETARUNS, in its deep version applicable to closed domains, that is to say domains for which the lexical semantics is fully specified and does not have to be induced. In addition, no ontology is needed: semantic re- lations are derived from linguistic relations encoded in the syntax. The main tenet of the system is that it is possible to produce consistent seman- tic representations using a strict linguistic approach without resorting to extralinguistic knowledge sources. The paper will briefly present the low level component which is responsible for pronominal binding, quantifier raising and temporal interpretation. Then it will discuss in more detail the high level component where a Discourse Model is created from text. The system has been evaluated on a wide variety of texts from closed domains, producing full and accurate parsing, semantics and anaphora resolution for all sentences.File | Dimensione | Formato | |
---|---|---|---|
W08-2209.pdf
non disponibili
Tipologia:
Abstract
Licenza:
Licenza non definita
Dimensione
98.22 kB
Formato
Adobe PDF
|
98.22 kB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.