Thunder-DeID: Accurate and Efficient De-identification Framework for Korean Court Judgments

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hahm, Sungeun, Kim, Heejin, Lee, Gyuseong, Park, Hyunji, Lee, Jaejin
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909849024987136
author Hahm, Sungeun
Kim, Heejin
Lee, Gyuseong
Park, Hyunji
Lee, Jaejin
author_facet Hahm, Sungeun
Kim, Heejin
Lee, Gyuseong
Park, Hyunji
Lee, Jaejin
contents To ensure a balance between open access to justice and personal data protection, the South Korean judiciary mandates the de-identification of court judgments before they can be publicly disclosed. However, the current de-identification process is inadequate for handling court judgments at scale while adhering to strict legal requirements. Additionally, the legal definitions and categorizations of personal identifiers are vague and not well-suited for technical solutions. To tackle these challenges, we propose a de-identification framework called Thunder-DeID, which aligns with relevant laws and practices. Specifically, we (i) construct and release the first Korean legal dataset containing annotated judgments along with corresponding lists of entity mentions, (ii) introduce a systematic categorization of Personally Identifiable Information (PII), and (iii) develop an end-to-end deep neural network (DNN)-based de-identification pipeline. Our experimental results demonstrate that our model achieves state-of-the-art performance in the de-identification of court judgments.
format Preprint
id arxiv_https___arxiv_org_abs_2506_15266
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Thunder-DeID: Accurate and Efficient De-identification Framework for Korean Court Judgments
Hahm, Sungeun
Kim, Heejin
Lee, Gyuseong
Park, Hyunji
Lee, Jaejin
Computation and Language
To ensure a balance between open access to justice and personal data protection, the South Korean judiciary mandates the de-identification of court judgments before they can be publicly disclosed. However, the current de-identification process is inadequate for handling court judgments at scale while adhering to strict legal requirements. Additionally, the legal definitions and categorizations of personal identifiers are vague and not well-suited for technical solutions. To tackle these challenges, we propose a de-identification framework called Thunder-DeID, which aligns with relevant laws and practices. Specifically, we (i) construct and release the first Korean legal dataset containing annotated judgments along with corresponding lists of entity mentions, (ii) introduce a systematic categorization of Personally Identifiable Information (PII), and (iii) develop an end-to-end deep neural network (DNN)-based de-identification pipeline. Our experimental results demonstrate that our model achieves state-of-the-art performance in the de-identification of court judgments.
title Thunder-DeID: Accurate and Efficient De-identification Framework for Korean Court Judgments
topic Computation and Language
url https://arxiv.org/abs/2506.15266