Question Answering (QA) over Knowledge Graphs (KGs) has gained its momentum thanks to the spread of the Semantic Web. However, despite the abundance of methods proposed in this field, there are still many aspects that need to be fully covered. One of them is the generation of SPARQL queries with modifiers, i.e. queries that are made up not only by triple patterns but also other terms belonging to the SPARQL syntax, such as FILTER, LIMIT, COUNT, ORDER BY. This task results difficult to accomplish in a generic way since the matching with natural language is not straightforward. Few works try to address this complex issue. In this paper, we propose a new approach to handle and to generate queries containing modifiers. Our method is able to generate queries with multiple modifiers, it is easily extendable to cover new modifiers and new languages, and it is independent of the KG structure. Our approach represents an extension of an existing work called QAnswer.

Handling Modifiers in Question Answering over Knowledge Graphs

Siciliani L.;Basile P.;Lops P.
2019

Abstract

Question Answering (QA) over Knowledge Graphs (KGs) has gained its momentum thanks to the spread of the Semantic Web. However, despite the abundance of methods proposed in this field, there are still many aspects that need to be fully covered. One of them is the generation of SPARQL queries with modifiers, i.e. queries that are made up not only by triple patterns but also other terms belonging to the SPARQL syntax, such as FILTER, LIMIT, COUNT, ORDER BY. This task results difficult to accomplish in a generic way since the matching with natural language is not straightforward. Few works try to address this complex issue. In this paper, we propose a new approach to handle and to generate queries containing modifiers. Our method is able to generate queries with multiple modifiers, it is easily extendable to cover new modifiers and new languages, and it is independent of the KG structure. Our approach represents an extension of an existing work called QAnswer.
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.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11586/272246
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact