Anonymization of Court Orders

Claus Povlsen; Bart Jongejan; Dorte Haltrup Hansen; Bo Krantz Simonsen

doi:10.1109/CISTI.2016.7521611

Anonymization of Court Orders

Bidragets oversatte titel: Anonymisering af Dommerkendelser

Claus Povlsen, Bart Jongejan, Dorte Haltrup Hansen, Bo Krantz Simonsen

Institut for Nordiske Studier og Sprogvidenskab

Publikation: Konferencebidrag › Paper › Forskning › peer review

4 Citationer (Scopus)

Abstract

We describe an anonymization tool that was commissioned by and specified together with Schultz, a publishing company specialized in Danish law related publications. Unavailability of training data and the need to guarantee compliance with pre-existing anonymization guidelines forced us to implement a tool using manually crafted rules. We used Bracmat, a programming language that is specialized in transforming tree data structures, to meet the requirement to pass the XML structure of the input document unscathed through the whole workflow. The tool attains a reassuringly good recall, makes almost no chunk errors and reduces the found entity designators to a nearly correct set of entities that the input text refers to, minimizing the time needed for manual check and post-editing.

Bidragets oversatte titel	Anonymisering af Dommerkendelser
Originalsprog	Engelsk
Publikationsdato	jun. 2016
Antal sider	4
DOI	https://doi.org/10.1109/CISTI.2016.7521611
Status	Udgivet - jun. 2016
Begivenhed	Iberian Conference on Information Systems and Technologies - Gran Canaria, Spanien Varighed: 15 jun. 2016 → 18 jun. 2016 Konferencens nummer: 11

Konference

Konference	Iberian Conference on Information Systems and Technologies
Nummer	11
Lokation	Gran Canaria
Land/Område	Spanien
Periode	15/06/2016 → 18/06/2016

Emneord

Det Humanistiske Fakultet
Named Entity Recognition
consistent assignment
high recall rate
real life application

Adgang til dokumentet

10.1109/CISTI.2016.7521611Licens: Ikke-specificeret

Automatisk anonymisering af fortrolige dokumenter:
Claus Povlsen (Foredragsholder)
14 aug. 2012
Aktivitet: Tale eller præsentation - typer › Foredrag og mundtlige bidrag

Citationsformater

@conference{5259240224d1469890230eff9919e0a7,

title = "Anonymization of Court Orders",

abstract = "We describe an anonymization tool that was commissioned by and specified together with Schultz, a publishing company specialized in Danish law related publications. Unavailability of training data and the need to guarantee compliance with pre-existing anonymization guidelines forced us to implement a tool using manually crafted rules. We used Bracmat, a programming language that is specialized in transforming tree data structures, to meet the requirement to pass the XML structure of the input document unscathed through the whole workflow. The tool attains a reassuringly good recall, makes almost no chunk errors and reduces the found entity designators to a nearly correct set of entities that the input text refers to, minimizing the time needed for manual check and post-editing.",

keywords = "Faculty of Humanities, Named Entity Recognition, consistent assignment, high recall rate, real life application",

author = "Claus Povlsen and Bart Jongejan and Hansen, {Dorte Haltrup} and {Krantz Simonsen}, Bo",

year = "2016",

month = jun,

doi = "10.1109/CISTI.2016.7521611",

language = "English",

note = "null ; Conference date: 15-06-2016 Through 18-06-2016",

}

TY - CONF

T1 - Anonymization of Court Orders

AU - Povlsen, Claus

AU - Jongejan, Bart

AU - Hansen, Dorte Haltrup

AU - Krantz Simonsen, Bo

N1 - Conference code: 11

PY - 2016/6

Y1 - 2016/6

N2 - We describe an anonymization tool that was commissioned by and specified together with Schultz, a publishing company specialized in Danish law related publications. Unavailability of training data and the need to guarantee compliance with pre-existing anonymization guidelines forced us to implement a tool using manually crafted rules. We used Bracmat, a programming language that is specialized in transforming tree data structures, to meet the requirement to pass the XML structure of the input document unscathed through the whole workflow. The tool attains a reassuringly good recall, makes almost no chunk errors and reduces the found entity designators to a nearly correct set of entities that the input text refers to, minimizing the time needed for manual check and post-editing.

AB - We describe an anonymization tool that was commissioned by and specified together with Schultz, a publishing company specialized in Danish law related publications. Unavailability of training data and the need to guarantee compliance with pre-existing anonymization guidelines forced us to implement a tool using manually crafted rules. We used Bracmat, a programming language that is specialized in transforming tree data structures, to meet the requirement to pass the XML structure of the input document unscathed through the whole workflow. The tool attains a reassuringly good recall, makes almost no chunk errors and reduces the found entity designators to a nearly correct set of entities that the input text refers to, minimizing the time needed for manual check and post-editing.

KW - Faculty of Humanities

KW - Named Entity Recognition

KW - consistent assignment

KW - high recall rate

KW - real life application

U2 - 10.1109/CISTI.2016.7521611

DO - 10.1109/CISTI.2016.7521611

M3 - Paper

Y2 - 15 June 2016 through 18 June 2016

ER -

Anonymization of Court Orders

Abstract

Konference

Emneord

Adgang til dokumentet

Aktiviteter

Automatisk anonymisering af fortrolige dokumenter:

Citationsformater