Measuring the relative performance of schema matchers

Shlomo Berkovsky; Yaniv Eytani; Avigdor Gal

doi:10.1109/WI.2005.94

Measuring the relative performance of schema matchers

Shlomo Berkovsky, Yaniv Eytani, Avigdor Gal

Data and Decision Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Schema matching is a complex process focusing on matching between concepts describing the data in heterogeneous data sources. There is a shift from manual schema matching, done by human experts, to automatic matching, using various heuristics (schema matchers). In this work, we consider the problem of linearly combining the results of a set of schema matchers. We propose the use of machine learning algorithms to learn the optimal weight assignments, given a set of schema matchers. We also suggest the use of genetic algorithms to improve the process efficiency.

Original language	English
Title of host publication	Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005
Pages	366-371
Number of pages	6
DOIs	https://doi.org/10.1109/WI.2005.94
State	Published - 2005
Event	2005 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2005 - Compiegne Cedex, France Duration: 19 Sep 2005 → 22 Sep 2005

Publication series

Name	Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005
Volume	2005

Conference

Conference	2005 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2005
Country/Territory	France
City	Compiegne Cedex,
Period	19/09/05 → 22/09/05

ASJC Scopus subject areas

General Engineering

Access to Document

10.1109/WI.2005.94

Cite this

@inproceedings{8195e2f565294190867df16a1a0098a7,

title = "Measuring the relative performance of schema matchers",

abstract = "Schema matching is a complex process focusing on matching between concepts describing the data in heterogeneous data sources. There is a shift from manual schema matching, done by human experts, to automatic matching, using various heuristics (schema matchers). In this work, we consider the problem of linearly combining the results of a set of schema matchers. We propose the use of machine learning algorithms to learn the optimal weight assignments, given a set of schema matchers. We also suggest the use of genetic algorithms to improve the process efficiency.",

author = "Shlomo Berkovsky and Yaniv Eytani and Avigdor Gal",

year = "2005",

doi = "10.1109/WI.2005.94",

language = "אנגלית",

isbn = "076952415X",

series = "Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005",

pages = "366--371",

booktitle = "Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005",

note = "2005 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2005 ; Conference date: 19-09-2005 Through 22-09-2005",

}

Berkovsky, S, Eytani, Y & Gal, A 2005, Measuring the relative performance of schema matchers. in Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005., 1517873, Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005, vol. 2005, pp. 366-371, 2005 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2005, Compiegne Cedex, France, 19/09/05. https://doi.org/10.1109/WI.2005.94

Measuring the relative performance of schema matchers. / Berkovsky, Shlomo; Eytani, Yaniv; Gal, Avigdor.
Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005. 2005. p. 366-371 1517873 (Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005; Vol. 2005).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Measuring the relative performance of schema matchers

AU - Berkovsky, Shlomo

AU - Eytani, Yaniv

AU - Gal, Avigdor

PY - 2005

Y1 - 2005

N2 - Schema matching is a complex process focusing on matching between concepts describing the data in heterogeneous data sources. There is a shift from manual schema matching, done by human experts, to automatic matching, using various heuristics (schema matchers). In this work, we consider the problem of linearly combining the results of a set of schema matchers. We propose the use of machine learning algorithms to learn the optimal weight assignments, given a set of schema matchers. We also suggest the use of genetic algorithms to improve the process efficiency.

AB - Schema matching is a complex process focusing on matching between concepts describing the data in heterogeneous data sources. There is a shift from manual schema matching, done by human experts, to automatic matching, using various heuristics (schema matchers). In this work, we consider the problem of linearly combining the results of a set of schema matchers. We propose the use of machine learning algorithms to learn the optimal weight assignments, given a set of schema matchers. We also suggest the use of genetic algorithms to improve the process efficiency.

UR - http://www.scopus.com/inward/record.url?scp=33748849857&partnerID=8YFLogxK

U2 - 10.1109/WI.2005.94

DO - 10.1109/WI.2005.94

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:33748849857

SN - 076952415X

SN - 9780769524153

T3 - Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005

SP - 366

EP - 371

BT - Proceedings - 2005 IEEE/WIC/ACM InternationalConference on Web Intelligence, WI 2005

T2 - 2005 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2005

Y2 - 19 September 2005 through 22 September 2005

ER -

Measuring the relative performance of schema matchers

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this