Title :
The retrieval system based on concept extending
Author :
Zhang, Maoyuan ; Liu, Ming
Author_Institution :
Dept. of Comput. Sci., HuaZhong Normal Univ., Wuhan, China
Abstract :
At present, the retrieval approaches could not match the concept with identical meaning but dissimilar expression. To aim at solving the problem, this paper puts forward a retrieval model based on concept extending. The system model is composed of three levels, which are the query level, the concept extending level, and the retrieval level. In the model, an extending method based on concept is proposed to extend user´s query from the concept hierarchy. This paper not only describes its extending algorithm after presenting a concept similarity between words, but also analyzes the merits of the extending method. Moreover, a result ranking method based on concept is proposed, to filter out the results with low scores at the concept level. In addition, the experiment´s result shows that the concept extending approach has improved the performance of retrieval system.
Keywords :
information filtering; query processing; concept extending; query level; result ranking method; results filtering; retrieval system; Algorithm design and analysis; Application software; Computational intelligence; Computer industry; Computer science; Content based retrieval; Filters; Fuzzy set theory; Humans; Information retrieval; concept extending; ranking; similarity;
Conference_Titel :
Computational Intelligence and Industrial Applications, 2009. PACIIA 2009. Asia-Pacific Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-4606-3
DOI :
10.1109/PACIIA.2009.5406582