Keyword query to graph query is a problem which aims at mapping entities and relationships mentioned in a query to entities and relationships in a knowledge-base. It has a very important application in different domains like that of information retrieval, data mining, genomics etc. Obviously keyword queries are preferred by users because they are based on natural language. Even though this point is a plus from the users point of view, it is a challenge from the machines perspective. This is because of the fact that natural languages are ambiguous and keyword queries are over specified. In this work a model decomposed into three sub-components is proposed. The first sub-component is used to remove ambiguity from the keywords and map them to clearly known entities in a knowledge base. The remaining two sub-components are used to extract interesting relationships between the clearly known entities. At the end top-k output sub-graphs are returned as the final output of the model. These results show mappings of the keywords to entities and the relationship between the mapped entities. Experimental results proved that the model has a good performance in terms of both quality and efficiency. Particularly the quality is similar to what is reported in state-of-the-art works. The experiments are performed on both synthetic and real queries.

Keyword query to graph query

Mottin, Davide;Palpanas, Themistoklis
2013

Abstract

Keyword query to graph query is a problem which aims at mapping entities and relationships mentioned in a query to entities and relationships in a knowledge-base. It has a very important application in different domains like that of information retrieval, data mining, genomics etc. Obviously keyword queries are preferred by users because they are based on natural language. Even though this point is a plus from the users point of view, it is a challenge from the machines perspective. This is because of the fact that natural languages are ambiguous and keyword queries are over specified. In this work a model decomposed into three sub-components is proposed. The first sub-component is used to remove ambiguity from the keywords and map them to clearly known entities in a knowledge base. The remaining two sub-components are used to extract interesting relationships between the clearly known entities. At the end top-k output sub-graphs are returned as the final output of the model. These results show mappings of the keywords to entities and the relationship between the mapped entities. Experimental results proved that the model has a good performance in terms of both quality and efficiency. Particularly the quality is similar to what is reported in state-of-the-art works. The experiments are performed on both synthetic and real queries.
Trento
Università degli Studi di Trento
Z., Kefato; M., Lissandrini; Mottin, Davide; Palpanas, Themistoklis
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/11572/101769
 Attenzione

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

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