FairER: Entity Resolution with Fairness Constraints

Vasilis Efthymiou; Kostas Stefanidis; Evaggelia Pitoura; Vassilis Christophides

doi:10.1145/3459637.3482105

FairER: Entity Resolution with Fairness Constraints

Vasilis Efthymiou
, Kostas Stefanidis
, Evaggelia Pitoura
, Vassilis Christophides

Tietotekniikka

Tutkimustuotos: Konferenssiartikkeli › Tieteellinen › vertaisarvioitu

20 Sitaatiot (Scopus)

48 Lataukset (Pure)

Abstrakti

There is an urgent call to detect and prevent "biased data"at the earliest possible stage of the data pipelines used to build automated decision-making systems. In this paper, we are focusing on controlling the data bias in entity resolution (ER) tasks aiming to discover and unify records/descriptions from different data sources that refer to the same real-world entity. We formally define the ER problem with fairness constraints ensuring that all groups of entities have similar chances to be resolved. Then, we introduce FairER, a greedy algorithm for solving this problem for fairness criteria based on equal matching decisions. Our experiments show that FairER achieves similar or higher accuracy against two baseline methods over 7 datasets, while guaranteeing minimal bias.

Alkuperäiskieli	Englanti
Otsikko	CIKM '21
Alaotsikko	Proceedings of the 30th ACM International Conference on Information and Knowledge Management
Kustantaja	ACM
Sivut	3004-3008
Sivumäärä	5
ISBN (elektroninen)	9781450384469
DOI - pysyväislinkit	https://doi.org/10.1145/3459637.3482105
Tila	Julkaistu - 26 lokak. 2021
OKM-julkaisutyyppi	A4 Artikkeli konferenssijulkaisussa
Tapahtuma	ACM International Conference on Information and Knowledge Management - Kesto: 1 tammik. 2013 → …

Julkaisusarja

Nimi	ACM International Conference on Information & Knowledge Management
ISSN (painettu)	2155-0751

Conference

Conference	ACM International Conference on Information and Knowledge Management
Ajanjakso	1/01/13 → …

Rahoitus

This project has received funding from the Hellenic Foundation for Research and Innovation (HFRI) and the General Secretariat for Research and Technology (GSRT), under grant agreement No 969.

Julkaisufoorumi-taso

Jufo-taso 2

!!ASJC Scopus subject areas

Yleinen liiketoiminta, johto ja kirjanpito
Yleinen päätöstiede

Pääsy asiakirjaan

10.1145/3459637.3482105

cikm2021
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Accepted author manuscript, 622 KBLisenssi: Unspecified

https://urn.fi/URN:NBN:fi:tuni-202210267881Lisenssi: Unspecified

Siteeraa tätä

Efthymiou, V., Stefanidis, K., Pitoura, E., & Christophides, V. (2021). FairER: Entity Resolution with Fairness Constraints. teoksessa CIKM '21: Proceedings of the 30th ACM International Conference on Information and Knowledge Management (Sivut 3004-3008). (ACM International Conference on Information & Knowledge Management). ACM. https://doi.org/10.1145/3459637.3482105

@inproceedings{d271534915214cbc8cfb0aa9bb3f99b0,

title = "FairER: Entity Resolution with Fairness Constraints",

abstract = "There is an urgent call to detect and prevent {"}biased data{"}at the earliest possible stage of the data pipelines used to build automated decision-making systems. In this paper, we are focusing on controlling the data bias in entity resolution (ER) tasks aiming to discover and unify records/descriptions from different data sources that refer to the same real-world entity. We formally define the ER problem with fairness constraints ensuring that all groups of entities have similar chances to be resolved. Then, we introduce FairER, a greedy algorithm for solving this problem for fairness criteria based on equal matching decisions. Our experiments show that FairER achieves similar or higher accuracy against two baseline methods over 7 datasets, while guaranteeing minimal bias.",

keywords = "algorithmic fairness, entity resolution",

author = "Vasilis Efthymiou and Kostas Stefanidis and Evaggelia Pitoura and Vassilis Christophides",

note = "Funding Information: This project has received funding from the Hellenic Foundation for Research and Innovation (HFRI) and the General Secretariat for Research and Technology (GSRT), under grant agreement No 969. Publisher Copyright: {\textcopyright} 2021 ACM. JUFOID=50076; ACM International Conference on Information and Knowledge Management ; Conference date: 01-01-2013",

year = "2021",

month = oct,

day = "26",

doi = "10.1145/3459637.3482105",

language = "English",

series = "ACM International Conference on Information \& Knowledge Management",

publisher = "ACM",

pages = "3004--3008",

booktitle = "CIKM '21",

address = "United States",

}

Efthymiou, V, Stefanidis, K, Pitoura, E & Christophides, V 2021, FairER: Entity Resolution with Fairness Constraints. julkaisussa CIKM '21: Proceedings of the 30th ACM International Conference on Information and Knowledge Management. ACM International Conference on Information & Knowledge Management, ACM, Sivut 3004-3008, ACM International Conference on Information and Knowledge Management, 1/01/13. https://doi.org/10.1145/3459637.3482105

FairER: Entity Resolution with Fairness Constraints. / Efthymiou, Vasilis; Stefanidis, Kostas; Pitoura, Evaggelia et al.
CIKM '21: Proceedings of the 30th ACM International Conference on Information and Knowledge Management. ACM, 2021. s. 3004-3008 (ACM International Conference on Information & Knowledge Management).

Tutkimustuotos: Konferenssiartikkeli › Tieteellinen › vertaisarvioitu

TY - GEN

T1 - FairER

T2 - ACM International Conference on Information and Knowledge Management

AU - Efthymiou, Vasilis

AU - Stefanidis, Kostas

AU - Pitoura, Evaggelia

AU - Christophides, Vassilis

N1 - Funding Information: This project has received funding from the Hellenic Foundation for Research and Innovation (HFRI) and the General Secretariat for Research and Technology (GSRT), under grant agreement No 969. Publisher Copyright: © 2021 ACM. JUFOID=50076

PY - 2021/10/26

Y1 - 2021/10/26

N2 - There is an urgent call to detect and prevent "biased data"at the earliest possible stage of the data pipelines used to build automated decision-making systems. In this paper, we are focusing on controlling the data bias in entity resolution (ER) tasks aiming to discover and unify records/descriptions from different data sources that refer to the same real-world entity. We formally define the ER problem with fairness constraints ensuring that all groups of entities have similar chances to be resolved. Then, we introduce FairER, a greedy algorithm for solving this problem for fairness criteria based on equal matching decisions. Our experiments show that FairER achieves similar or higher accuracy against two baseline methods over 7 datasets, while guaranteeing minimal bias.

AB - There is an urgent call to detect and prevent "biased data"at the earliest possible stage of the data pipelines used to build automated decision-making systems. In this paper, we are focusing on controlling the data bias in entity resolution (ER) tasks aiming to discover and unify records/descriptions from different data sources that refer to the same real-world entity. We formally define the ER problem with fairness constraints ensuring that all groups of entities have similar chances to be resolved. Then, we introduce FairER, a greedy algorithm for solving this problem for fairness criteria based on equal matching decisions. Our experiments show that FairER achieves similar or higher accuracy against two baseline methods over 7 datasets, while guaranteeing minimal bias.

KW - algorithmic fairness

KW - entity resolution

U2 - 10.1145/3459637.3482105

DO - 10.1145/3459637.3482105

M3 - Conference contribution

AN - SCOPUS:85119203325

T3 - ACM International Conference on Information & Knowledge Management

SP - 3004

EP - 3008

BT - CIKM '21

PB - ACM

Y2 - 1 January 2013

ER -

FairER: Entity Resolution with Fairness Constraints

Abstrakti

Julkaisusarja

Conference

Rahoitus

Julkaisufoorumi-taso

!!ASJC Scopus subject areas

Pääsy asiakirjaan

Sormenjälki

Siteeraa tätä