Please use this identifier to cite or link to this item:
|Title:||From structure-based to semantics-based: Towards effective XML keyword search|
|Source:||Le, T.N.,Wu, H.,Ling, T.W.,Li, L.,Lu, J. (2013). From structure-based to semantics-based: Towards effective XML keyword search. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 8217 LNCS : 356-371. ScholarBank@NUS Repository. https://doi.org/10.1007/978-3-642-41924-9_29|
|Abstract:||Existing XML keyword search approaches can be categorized into tree-based search and graph-based search. Both of them are structure-based search because they mainly rely on the exploration of the structural features of document. Those structure-based approaches cannot fully exploit hidden semantics in XML document. This causes serious problems in processing some class of keyword queries. In this paper, we thoroughly point out mismatches between answers returned by structure-based search and the expectations of common users. Through detailed analysis of these mismatches, we show the importance of semantics in XML keyword search and propose a semantics-based approach to process XML keyword queries. Particularly, we propose to use Object Relationship (OR) graph, which fully captures semantics of object, relationship and attribute, to represent XML document and we develop algorithms based on the OR graph to return more comprehensive answers. Experimental results show that our proposed semantics-based approach can resolve the problems of the structure-based search, and significantly improve both the effectiveness and efficiency. © Springer-Verlag 2013.|
|Source Title:||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Feb 20, 2018
checked on Feb 16, 2018
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.