New Weighting Schemes for Meta-blocking

Nikolaus Augsten; Roland Kwitt; Matteo  Lissandrini; Willi Mann; Themis  Palpanas; George  Papadakis

New Weighting Schemes for Meta-blocking

Nikolaus Augsten, Roland Kwitt, Matteo Lissandrini, Willi Mann, Themis Palpanas, George Papadakis

Publikation: Andere Beiträge › Sonstiger Beitrag › Forschung

Abstract

Entity Resolution constitutes a core data integration task that relies on Blocking in order to tame its quadratic time complexity. Schema-agnostic blocking comes at the cost of many irrelevant candidate pairs (i.e., comparisons), which can be significantly reduced with Meta-blocking. In Meta-blocking, a weighting scheme is first applied on every pair of candidate entities in proportion to the likelihood that they are matching, and a pruning algorithm then discards the pairs with the lowest scores.
In this work, we briefly discuss the existing Meta-blocking weighting schemes, and then propose four new weighting schemes that can be used by Meta-blocking techniques.

Originalsprache	Englisch
Publikationsstatus	Veröffentlicht - 1 Okt. 2021

Systematik der Wissenschaftszweige 2012

102 Informatik

Zugriff auf Dokument

http://lipade.mi.parisdescartes.fr/?page_id=611

1 Konferenzbeitrag

Scaling Density-Based Clustering to Large Collections of Sets
Kocher, D., Mann, W. & Augsten, N., März 2021, EDBT 2021 - Proceedings of the 24th International Conference on Extending Database Technology: 24th International Conference on Extending Database Technology, Proceedings. Velegrakis, Y., Velegrakis, Y., Zeinalipour, D., Chrysanthis, P. K., Chrysanthis, P. K. & Guerra, F. (Hrsg.). OpenProceedings.org, S. 109-120 12 S. (Advances in Database Technology - EDBT; Band 2021-March).
Publikation: Beitrag in Buch/Bericht/Konferenzband/Gesetzeskommentar › Konferenzbeitrag › Peer-reviewed

Open Access

Dieses zitieren

@misc{211f5a5d9a3f4921ac82ece0d5b29bcf,

title = "New Weighting Schemes for Meta-blocking",

abstract = "Entity Resolution constitutes a core data integration task that relies on Blocking in order to tame its quadratic time complexity. Schema-agnostic blocking comes at the cost of many irrelevant candidate pairs (i.e., comparisons), which can be significantly reduced with Meta-blocking. In Meta-blocking, a weighting scheme is first applied on every pair of candidate entities in proportion to the likelihood that they are matching, and a pruning algorithm then discards the pairs with the lowest scores.In this work, we briefly discuss the existing Meta-blocking weighting schemes, and then propose four new weighting schemes that can be used by Meta-blocking techniques.",

author = "Nikolaus Augsten and Roland Kwitt and Matteo Lissandrini and Willi Mann and Themis Palpanas and George Papadakis",

year = "2021",

month = oct,

day = "1",

language = "English",

type = "Other",

}

TY - GEN

T1 - New Weighting Schemes for Meta-blocking

AU - Augsten, Nikolaus

AU - Kwitt, Roland

AU - Lissandrini, Matteo

AU - Mann, Willi

AU - Palpanas, Themis

AU - Papadakis, George

PY - 2021/10/1

Y1 - 2021/10/1

N2 - Entity Resolution constitutes a core data integration task that relies on Blocking in order to tame its quadratic time complexity. Schema-agnostic blocking comes at the cost of many irrelevant candidate pairs (i.e., comparisons), which can be significantly reduced with Meta-blocking. In Meta-blocking, a weighting scheme is first applied on every pair of candidate entities in proportion to the likelihood that they are matching, and a pruning algorithm then discards the pairs with the lowest scores.In this work, we briefly discuss the existing Meta-blocking weighting schemes, and then propose four new weighting schemes that can be used by Meta-blocking techniques.

AB - Entity Resolution constitutes a core data integration task that relies on Blocking in order to tame its quadratic time complexity. Schema-agnostic blocking comes at the cost of many irrelevant candidate pairs (i.e., comparisons), which can be significantly reduced with Meta-blocking. In Meta-blocking, a weighting scheme is first applied on every pair of candidate entities in proportion to the likelihood that they are matching, and a pruning algorithm then discards the pairs with the lowest scores.In this work, we briefly discuss the existing Meta-blocking weighting schemes, and then propose four new weighting schemes that can be used by Meta-blocking techniques.

M3 - Other contribution

ER -

New Weighting Schemes for Meta-blocking

Abstract

Systematik der Wissenschaftszweige 2012

Zugriff auf Dokument

Publikationen

Scaling Density-Based Clustering to Large Collections of Sets

Dieses zitieren