Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A hybrid solution of ontology‐based query expansion

A hybrid solution of ontology‐based query expansion Purpose – The purpose of this paper is to propose a hybrid ontology‐based solution to expand user's queries. Design/methodology/approach – The solution aims for ontology development and query expansion with ontology‐based approach. The first task is to develop an ontology (named OMP), which relates to key‐properties and key‐members of objects described in words/terms of English vocabulary. Its training methodology is also a hybrid, rule‐based with proposed patterns and statistical‐based solution for selecting the best candidates from TREC English corpus. The second is proposals for mechanisms not only to look for relative result in the ontology OMP to complete and expand user's entered query/noun phrase, but also to expand the search progress by linking the OMP ontology to indexes of information retrieval system. Especially, the base of these two tasks is our proposal for four kinds of semantic relationship of words. Findings – Several semantic relationships among words in vocabulary has been introduced and currently used in WordNet to represent the system of semantic networks. In another way, our analyzing for words in English vocabulary found that there are some kinds of semantic dependency in some cases for part(s) of a noun phrase, and it can be represented in grammar noun phrase syntax. That affects not only the proposed approach of ontology OMP development via identifying four kinds of semantic relationship and organizing its structure including core element types such as object and key‐member and key‐property, but also ontology training mechanism and solutions of query expansion by adding extended correspondent words (based on that relationship) to original query. Research limitations/implications – In initial iteration, the approach is applied for English query only with limited size of ontology OMP and dependency on grammar rules‐based in creating patterns to extract data from corpus. For future research, applications for other languages (Vietnamese, Chinese …) with sharp focus on improvement of ontology training quality/quantity and query expansion precision are primary targets. Practical implications – The developed ontology OMP can be shared as a support for other applications such as semantic data extraction or semantic information retrieval in other researches. Originality/value – This paper fulfils an approach of ontology‐based query expansion and theoretical definitions of semantic relationship among words. Specially, these kinds of relationship can use to develop a useful semantic network system. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Web Information Systems Emerald Publishing

A hybrid solution of ontology‐based query expansion

Loading next page...
 
/lp/emerald-publishing/a-hybrid-solution-of-ontology-based-query-expansion-zgUM42lrrT
Publisher
Emerald Publishing
Copyright
Copyright © 2008 Emerald Group Publishing Limited. All rights reserved.
ISSN
1744-0084
DOI
10.1108/17440080810882388
Publisher site
See Article on Publisher Site

Abstract

Purpose – The purpose of this paper is to propose a hybrid ontology‐based solution to expand user's queries. Design/methodology/approach – The solution aims for ontology development and query expansion with ontology‐based approach. The first task is to develop an ontology (named OMP), which relates to key‐properties and key‐members of objects described in words/terms of English vocabulary. Its training methodology is also a hybrid, rule‐based with proposed patterns and statistical‐based solution for selecting the best candidates from TREC English corpus. The second is proposals for mechanisms not only to look for relative result in the ontology OMP to complete and expand user's entered query/noun phrase, but also to expand the search progress by linking the OMP ontology to indexes of information retrieval system. Especially, the base of these two tasks is our proposal for four kinds of semantic relationship of words. Findings – Several semantic relationships among words in vocabulary has been introduced and currently used in WordNet to represent the system of semantic networks. In another way, our analyzing for words in English vocabulary found that there are some kinds of semantic dependency in some cases for part(s) of a noun phrase, and it can be represented in grammar noun phrase syntax. That affects not only the proposed approach of ontology OMP development via identifying four kinds of semantic relationship and organizing its structure including core element types such as object and key‐member and key‐property, but also ontology training mechanism and solutions of query expansion by adding extended correspondent words (based on that relationship) to original query. Research limitations/implications – In initial iteration, the approach is applied for English query only with limited size of ontology OMP and dependency on grammar rules‐based in creating patterns to extract data from corpus. For future research, applications for other languages (Vietnamese, Chinese …) with sharp focus on improvement of ontology training quality/quantity and query expansion precision are primary targets. Practical implications – The developed ontology OMP can be shared as a support for other applications such as semantic data extraction or semantic information retrieval in other researches. Originality/value – This paper fulfils an approach of ontology‐based query expansion and theoretical definitions of semantic relationship among words. Specially, these kinds of relationship can use to develop a useful semantic network system.

Journal

International Journal of Web Information SystemsEmerald Publishing

Published: Jun 20, 2008

Keywords: Computer software; Query languages; Semantics

References