Efficient evaluation of distance predicates in Xpath full-text query

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In recent years, more and more XML repositories are emerging, e.g., XML digital library, SIGMOD and DBLP document collections. Since XML is good at representing both structured and unstructured data, to facilitate the usage of this kind of information, it is necessary to support structure-based and content-based (full-text) queries/retrievals over XML repositories. With existing XPath/XQuery Pull-Text, user could do search based on cardinality, proximity or distance predicates. In this paper, we propose an efficient approach for the Information Retrieval (IR) style search, especially distance predicates search, on XML documents. Numbering technique is employed to encode XML documents, and then three algorithms are designed to evaluate queries with distance predicates. To improve the performance, some optimization techniques are introduced. Extensive experiments show the effectiveness and efficiency of the proposed approach.

Original languageEnglish
Title of host publicationAdvanced Web and Network Technologies, and Applications - APWeb 2006 International Workshops
Subtitle of host publicationXRA, IWSN, MEGA, and ICSE, Proceedings
PublisherSpringer Verlag
Pages76-85
Number of pages10
ISBN (Print)3540311580, 9783540311584
DOIs
StatePublished - 2006
Externally publishedYes
EventAPWeb 2006 International Workshops: XRA, IWSN, MEGA, and ICSE - Harbin, China
Duration: 16 Jan 200618 Jan 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3842 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceAPWeb 2006 International Workshops: XRA, IWSN, MEGA, and ICSE
Country/TerritoryChina
CityHarbin
Period16/01/0618/01/06

Fingerprint

Dive into the research topics of 'Efficient evaluation of distance predicates in Xpath full-text query'. Together they form a unique fingerprint.

Cite this