Data di Pubblicazione:
2011
Citazione:
A Quad-Tree Based Multiresolution Approach for Two-dimensional Summary Data / Buccafurri, F., Furfaro, F., Mazzeo, G., Saccà, D.. - In: INFORMATION SYSTEMS. - ISSN 0306-4379. - 36(7):(2011), pp. 1082-1103.
Abstract:
Evaluatingaggregaterangequeriesbyaccessingacompressedrepresentationofthedataisa
widely adoptedsolutiontotheproblemofefficientlyretrievingaggregateinformationfrom
largeamountsofdata.Althoughseveralsummarizationtechniqueshavebeenproposed
which areeffectiveinreducingtheamount of timeneededforcomputingaggregates,
queryingsummarydataoftenresultsindramatically inaccurateestimates,duetothe
difficultyoflimitingthelossofinformationresultingfromdatacompression.Thus,acrucial
issueregardingthedefinitionofsummarizationtechniquesistoretainareasonabledegree
of approximationinreconstructingqueryanswers. Followingtheideathataneffective adhoc
solutiontothisproblemcanbefoundinspecific applicationdomains,inthispaperwe
restrictourattentiontothecaseoftwo-dimensionaldata,whichisrelevantforanumberof
applications.Ourproposalisasummarization techniquewhereblocksofdataresultingfrom
a quad-treebasedpartitionofthetwo-dimensionaldomainaresummarizedintoaggregate
values andpossiblyassociatedwith indices, i.e.,compactstructuresprovidinganapproximate
descriptiontheoriginaldatainsidethem.Severalexperimentalresultsarepresented
showing thatourtechniqueresultsindatasynopsesprovidingqueryestimateshavingerror
rateslowerthanothertechniquestailoredatdatawithagenericdimensionality,suchas
wavelets andvarioustypesofmulti-dimensionalhistogram.
widely adoptedsolutiontotheproblemofefficientlyretrievingaggregateinformationfrom
largeamountsofdata.Althoughseveralsummarizationtechniqueshavebeenproposed
which areeffectiveinreducingtheamount of timeneededforcomputingaggregates,
queryingsummarydataoftenresultsindramatically inaccurateestimates,duetothe
difficultyoflimitingthelossofinformationresultingfromdatacompression.Thus,acrucial
issueregardingthedefinitionofsummarizationtechniquesistoretainareasonabledegree
of approximationinreconstructingqueryanswers. Followingtheideathataneffective adhoc
solutiontothisproblemcanbefoundinspecific applicationdomains,inthispaperwe
restrictourattentiontothecaseoftwo-dimensionaldata,whichisrelevantforanumberof
applications.Ourproposalisasummarization techniquewhereblocksofdataresultingfrom
a quad-treebasedpartitionofthetwo-dimensionaldomainaresummarizedintoaggregate
values andpossiblyassociatedwith indices, i.e.,compactstructuresprovidinganapproximate
descriptiontheoriginaldatainsidethem.Severalexperimentalresultsarepresented
showing thatourtechniqueresultsindatasynopsesprovidingqueryestimateshavingerror
rateslowerthanothertechniquestailoredatdatawithagenericdimensionality,suchas
wavelets andvarioustypesofmulti-dimensionalhistogram.
Tipologia CRIS:
1.1 Articolo in rivista
Elenco autori:
Buccafurri, Francesco; Furfaro, F; Mazzeo, G; Saccà, D
Link alla scheda completa:
Pubblicato in: