PostMatch: A framework for efficient address matching

Darren Yates; Zahidul Islam; Yanchang Zhao; Richi Nayak; Vladimir Estivill-Castro; Salil Kanhere

doi:10.1007/978-981-16-8531-6_10

PostMatch: A framework for efficient address matching

Darren Yates, Zahidul Islam, Yanchang Zhao, Richi Nayak, Vladimir Estivill-Castro, Salil Kanhere

Research output: Book chapter/Published conference paper › Conference paper › peer-review

Abstract

Matching lists of addresses is an increasingly common task executed by business and governments alike. However, due to security issues, this task cannot always be performed using cloud computing.Moreover, addresses can arrive with spelling errors that can cause non-matches or ‘false negatives’ to occur. Our proposed framework, Post-Match, provides a locally-executed method for address-matching that combines the open-source ‘Libpostal’ address-parsing library with our ‘postparse’ post-processor code and machine-learning. PostMatch pro-vides improved parsing accuracy compared with Libpostal alone, approaching 96.9%. The matching process features the Jaro-Winkler edit distance algorithm together with XGBoost machine-learning to achiev every high accuracy on public data. PostMatch is open-source (GPL3 licensed) and available as R script code on Github.

Original language	English
Title of host publication	Data mining
Subtitle of host publication	19th Australasian conference on data mining, AusDM 2021, proceedings
Editors	Yue Xu, Anton Lord, Richi Nayak, Graham Williams, Rosalind Wang, Yee Ling Boo, Yanchang Zhao
Place of Publication	Singapore
Publisher	Springer
Pages	136-151
Number of pages	16
Volume	1504
ISBN (Electronic)	9789811685316
ISBN (Print)	9789811685309
DOIs	https://doi.org/10.1007/978-981-16-8531-6_10
Publication status	E-pub ahead of print - 09 Dec 2021
Event	19th Australasian Data Mining Conference 2021: (AusDM'21) - Online, Australia Duration: 14 Dec 2021 → 15 Dec 2021 https://ausdm21.ausdm.org/ https://link.springer.com/content/pdf/bfm%3A978-981-16-8531-6%2F1.pdf (Proceedings front matter) https://link.springer.com/book/10.1007/978-981-16-8531-6 (Proceedings)

Publication series

Name	Communications in Computer and Information Science
Publisher	Springer
Volume	1504
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	19th Australasian Data Mining Conference 2021
Country/Territory	Australia
Period	14/12/21 → 15/12/21
Other	The Australasian Data Mining Conference has established itself as the premier Australasian meeting for both practitioners and researchers in data mining. It is devoted to the art and science of intelligent analysis of (usually big) data sets for meaningful (and previously unknown) insights. This conference will enable the sharing and learning of research and progress in the local context and new breakthroughs in data mining algorithms and their applications across all industries.
Internet address	https://ausdm21.ausdm.org/ https://link.springer.com/content/pdf/bfm%3A978-981-16-8531-6%2F1.pdf (Proceedings front matter) https://link.springer.com/book/10.1007/978-981-16-8531-6 (Proceedings)

Access to Document

10.1007/978-981-16-8531-6_10

Cite this

Yates, D., Islam, Z., Zhao, Y., Nayak, R., Estivill-Castro, V., & Kanhere, S. (2021). PostMatch: A framework for efficient address matching. In Y. Xu, A. Lord, R. Nayak, G. Williams, R. Wang, Y. L. Boo, & Y. Zhao (Eds.), Data mining: 19th Australasian conference on data mining, AusDM 2021, proceedings (Vol. 1504, pp. 136-151). (Communications in Computer and Information Science; Vol. 1504). Springer. Advance online publication. https://doi.org/10.1007/978-981-16-8531-6_10

Yates, Darren ; Islam, Zahidul ; Zhao, Yanchang et al. / PostMatch : A framework for efficient address matching. Data mining: 19th Australasian conference on data mining, AusDM 2021, proceedings. editor / Yue Xu ; Anton Lord ; Richi Nayak ; Graham Williams ; Rosalind Wang ; Yee Ling Boo ; Yanchang Zhao. Vol. 1504 Singapore : Springer, 2021. pp. 136-151 (Communications in Computer and Information Science).

@inproceedings{873872c36b8f46e9aa8a698dc77b3945,

title = "PostMatch: A framework for efficient address matching",

abstract = "Matching lists of addresses is an increasingly common task executed by business and governments alike. However, due to security issues, this task cannot always be performed using cloud computing.Moreover, addresses can arrive with spelling errors that can cause non-matches or {\textquoteleft}false negatives{\textquoteright} to occur. Our proposed framework, Post-Match, provides a locally-executed method for address-matching that combines the open-source {\textquoteleft}Libpostal{\textquoteright} address-parsing library with our {\textquoteleft}postparse{\textquoteright} post-processor code and machine-learning. PostMatch pro-vides improved parsing accuracy compared with Libpostal alone, approaching 96.9%. The matching process features the Jaro-Winkler edit distance algorithm together with XGBoost machine-learning to achiev every high accuracy on public data. PostMatch is open-source (GPL3 licensed) and available as R script code on Github.",

keywords = "Address matching, Data matching, XGBoost",

author = "Darren Yates and Zahidul Islam and Yanchang Zhao and Richi Nayak and Vladimir Estivill-Castro and Salil Kanhere",

note = "Includes bibliographical references.; 19th Australasian Data Mining Conference 2021 : (AusDM'21) ; Conference date: 14-12-2021 Through 15-12-2021",

year = "2021",

month = dec,

day = "9",

doi = "10.1007/978-981-16-8531-6_10",

language = "English",

isbn = "9789811685309",

volume = "1504",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "136--151",

editor = "Yue Xu and Anton Lord and Richi Nayak and Williams, {Graham } and Rosalind Wang and Boo, {Yee Ling} and Yanchang Zhao",

booktitle = "Data mining",

address = "United States",

url = "https://ausdm21.ausdm.org/, https://link.springer.com/content/pdf/bfm%3A978-981-16-8531-6%2F1.pdf, https://link.springer.com/book/10.1007/978-981-16-8531-6",

}

Yates, D , Islam, Z, Zhao, Y, Nayak, R, Estivill-Castro, V & Kanhere, S 2021, PostMatch: A framework for efficient address matching. in Y Xu, A Lord, R Nayak, G Williams, R Wang, YL Boo & Y Zhao (eds), Data mining: 19th Australasian conference on data mining, AusDM 2021, proceedings. vol. 1504, Communications in Computer and Information Science, vol. 1504, Springer, Singapore, pp. 136-151, 19th Australasian Data Mining Conference 2021, Australia, 14/12/21. https://doi.org/10.1007/978-981-16-8531-6_10

PostMatch: A framework for efficient address matching. / Yates, Darren ; Islam, Zahidul; Zhao, Yanchang et al.
Data mining: 19th Australasian conference on data mining, AusDM 2021, proceedings. ed. / Yue Xu; Anton Lord; Richi Nayak; Graham Williams; Rosalind Wang; Yee Ling Boo; Yanchang Zhao. Vol. 1504 Singapore: Springer, 2021. p. 136-151 (Communications in Computer and Information Science; Vol. 1504).

Research output: Book chapter/Published conference paper › Conference paper › peer-review

TY - GEN

T1 - PostMatch

T2 - 19th Australasian Data Mining Conference 2021

AU - Yates, Darren

AU - Islam, Zahidul

AU - Zhao, Yanchang

AU - Nayak, Richi

AU - Estivill-Castro, Vladimir

AU - Kanhere, Salil

N1 - Includes bibliographical references.

PY - 2021/12/9

Y1 - 2021/12/9

N2 - Matching lists of addresses is an increasingly common task executed by business and governments alike. However, due to security issues, this task cannot always be performed using cloud computing.Moreover, addresses can arrive with spelling errors that can cause non-matches or ‘false negatives’ to occur. Our proposed framework, Post-Match, provides a locally-executed method for address-matching that combines the open-source ‘Libpostal’ address-parsing library with our ‘postparse’ post-processor code and machine-learning. PostMatch pro-vides improved parsing accuracy compared with Libpostal alone, approaching 96.9%. The matching process features the Jaro-Winkler edit distance algorithm together with XGBoost machine-learning to achiev every high accuracy on public data. PostMatch is open-source (GPL3 licensed) and available as R script code on Github.

AB - Matching lists of addresses is an increasingly common task executed by business and governments alike. However, due to security issues, this task cannot always be performed using cloud computing.Moreover, addresses can arrive with spelling errors that can cause non-matches or ‘false negatives’ to occur. Our proposed framework, Post-Match, provides a locally-executed method for address-matching that combines the open-source ‘Libpostal’ address-parsing library with our ‘postparse’ post-processor code and machine-learning. PostMatch pro-vides improved parsing accuracy compared with Libpostal alone, approaching 96.9%. The matching process features the Jaro-Winkler edit distance algorithm together with XGBoost machine-learning to achiev every high accuracy on public data. PostMatch is open-source (GPL3 licensed) and available as R script code on Github.

KW - Address matching

KW - Data matching

KW - XGBoost

U2 - 10.1007/978-981-16-8531-6_10

DO - 10.1007/978-981-16-8531-6_10

M3 - Conference paper

SN - 9789811685309

VL - 1504

T3 - Communications in Computer and Information Science

SP - 136

EP - 151

BT - Data mining

A2 - Xu, Yue

A2 - Lord, Anton

A2 - Nayak, Richi

A2 - Williams, Graham

A2 - Wang, Rosalind

A2 - Boo, Yee Ling

A2 - Zhao, Yanchang

PB - Springer

CY - Singapore

Y2 - 14 December 2021 through 15 December 2021

ER -

Yates D , Islam Z, Zhao Y, Nayak R, Estivill-Castro V, Kanhere S. PostMatch: A framework for efficient address matching. In Xu Y, Lord A, Nayak R, Williams G, Wang R, Boo YL, Zhao Y, editors, Data mining: 19th Australasian conference on data mining, AusDM 2021, proceedings. Vol. 1504. Singapore: Springer. 2021. p. 136-151. (Communications in Computer and Information Science). Epub 2021 Dec 9. doi: 10.1007/978-981-16-8531-6_10

PostMatch: A framework for efficient address matching

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this