Automated extraction of semantic information from German legal documents

Bernhard Waltl; Jörg Landthaler; Elena Scepankova; Florian Matthes; Thomas Geiger; Christoph Stocker; Christian Schneider

Automated extraction of semantic information from German legal documents

Bernhard Waltl
, Jörg Landthaler
, Elena Scepankova
, Florian Matthes
, Thomas Geiger
, Christoph Stocker
, Christian Schneider

Technical University of Munich
DATEV eG

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

Based on a collaborative data science environment, and a large document corpus (< 130'000 documents from German tax law) we demonstrate the extraction of semantic information. This paper shows the potential of rule-based text analysis to automatically extract semantic information, such as the year of dispute in cases. Additionally, it demonstrates the extraction of legal definitions in laws and the usage of terms in a defining context. Based on an iterative and interdisciplinary process, legal experts, software engineers, and data scientists evaluate and continuously refine the model used for the computer-supported extraction.

Original language	English
Journal	Jusletter IT
Issue number	February
State	Published - 23 Feb 2017

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 17 Partnerships for the Goals

Keywords

LegalData science
Semantic analysis
Structured information
Text mining

Cite this

@article{d240fc0e7c9f41c5bd6eca8478843c42,

title = "Automated extraction of semantic information from German legal documents",

abstract = "Based on a collaborative data science environment, and a large document corpus (< 130'000 documents from German tax law) we demonstrate the extraction of semantic information. This paper shows the potential of rule-based text analysis to automatically extract semantic information, such as the year of dispute in cases. Additionally, it demonstrates the extraction of legal definitions in laws and the usage of terms in a defining context. Based on an iterative and interdisciplinary process, legal experts, software engineers, and data scientists evaluate and continuously refine the model used for the computer-supported extraction.",

keywords = "LegalData science, Semantic analysis, Structured information, Text mining",

author = "Bernhard Waltl and J{\"o}rg Landthaler and Elena Scepankova and Florian Matthes and Thomas Geiger and Christoph Stocker and Christian Schneider",

year = "2017",

month = feb,

day = "23",

language = "English",

journal = "Jusletter IT",

issn = "1664-848X",

publisher = "Editions Weblaw",

number = "February",

}

TY - JOUR

T1 - Automated extraction of semantic information from German legal documents

AU - Waltl, Bernhard

AU - Landthaler, Jörg

AU - Scepankova, Elena

AU - Matthes, Florian

AU - Geiger, Thomas

AU - Stocker, Christoph

AU - Schneider, Christian

PY - 2017/2/23

Y1 - 2017/2/23

N2 - Based on a collaborative data science environment, and a large document corpus (< 130'000 documents from German tax law) we demonstrate the extraction of semantic information. This paper shows the potential of rule-based text analysis to automatically extract semantic information, such as the year of dispute in cases. Additionally, it demonstrates the extraction of legal definitions in laws and the usage of terms in a defining context. Based on an iterative and interdisciplinary process, legal experts, software engineers, and data scientists evaluate and continuously refine the model used for the computer-supported extraction.

AB - Based on a collaborative data science environment, and a large document corpus (< 130'000 documents from German tax law) we demonstrate the extraction of semantic information. This paper shows the potential of rule-based text analysis to automatically extract semantic information, such as the year of dispute in cases. Additionally, it demonstrates the extraction of legal definitions in laws and the usage of terms in a defining context. Based on an iterative and interdisciplinary process, legal experts, software engineers, and data scientists evaluate and continuously refine the model used for the computer-supported extraction.

KW - LegalData science

KW - Semantic analysis

KW - Structured information

KW - Text mining

UR - https://www.scopus.com/pages/publications/85037671993

M3 - Article

AN - SCOPUS:85037671993

SN - 1664-848X

JO - Jusletter IT

JF - Jusletter IT

IS - February

ER -

Automated extraction of semantic information from German legal documents

Abstract

UN SDGs

Keywords

Other files and links

Fingerprint

Cite this