Keywords filtering over probabilistic XML data

Chenjing Zhang, Le Chang, Chaofeng Sha, Xiaoling Wang, Aoying Zhou

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

5 Scopus citations

Abstract

Probabilistic XML data is widely used in many web applications. Recent work has been mostly focused on structured query over probabilistic XML data. A few of work has been done about keyword query. However only the independent and the mutually-exclusive relationship among sibling nodes are discussed. This paper addresses the problem of keyword filtering over probabilistic XML data, and we propose PrXML {exp, ind, mux} model to represent a more general relationship among XML sibling nodes, for keywords filtering over probabilistic XML data. kdptab is defined as keyword distribution probability table of one subtree. The Dot product, Cartesian product, and addition operation of kdptab are also defined. In PrXML {exp, ind, mux} model, XML document is scanned bottom-up and achieve keyword filtering based on SLCA semantics efficiently in our method. Finally, the features and efficiency of our method are evaluated with extensive experimental results.

Original languageEnglish
Title of host publicationWeb Technologies and Applications - 14th Asia-Pacific Web Conference, APWeb 2012, Proceedings
Pages183-194
Number of pages12
DOIs
StatePublished - 2012
Event14th Asia Pacific Web Technology Conference, APWeb 2012 - Kunming, China
Duration: 11 Apr 201213 Apr 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7235 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th Asia Pacific Web Technology Conference, APWeb 2012
Country/TerritoryChina
CityKunming
Period11/04/1213/04/12

Keywords

  • Keywords Filtering
  • Probabilistic XML
  • SLCA

Fingerprint

Dive into the research topics of 'Keywords filtering over probabilistic XML data'. Together they form a unique fingerprint.

Cite this