Yet Another Matcher - Archive ouverte HAL Access content directly
Reports Year : 2009

Yet Another Matcher

(1) , (2) , (2) , (3)
1
2
3
Remi Coletta
  • Function : Author
  • PersonId : 932759
Zohra Bellahsene
Renée J. Miller
  • Function : Author
  • PersonId : 861404

Abstract

Discovering correspondences between schema elements is a crucial task for data integration. Most matching tools are semi-automatic, e.g. an expert must tune some parameters (thresholds, weights, etc.). They mainly use several methods to combine and aggregate similarity measures. However, their quality results often decrease when one requires to integrate a new similarity measure or when matching particular domain schemas. This paper describes YAM (Yet Another Matcher), which is a matcher factory. Indeed, it enables the generation of a dedicated matcher for a given schema matching scenario, according to user inputs. Our approach is based on machine learning since schema matchers can be seen as classifiers. Several bunches of experiments run against matchers generated by YAM and traditional matching tools show how our approach (i) is able to generate the best matcher for a given scenario and (ii) easily integrates user preferences, namely recall and precision tradeoff.
Fichier principal
Vignette du fichier
CIKM2009_1245_53b101d9.pdf (271.39 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

lirmm-00399025 , version 1 (25-06-2010)

Identifiers

  • HAL Id : lirmm-00399025 , version 1

Cite

Fabien Duchateau, Remi Coletta, Zohra Bellahsene, Renée J. Miller. Yet Another Matcher. RR-09016, 2009. ⟨lirmm-00399025⟩
224 View
137 Download

Share

Gmail Facebook Twitter LinkedIn More