Extraction of Legal Definitions from a Japanese Statutory Corpus

impulseverseAI and Robotics

Oct 24, 2013 (3 years and 5 months ago)

215 views

Extraction of Legal Definitions from

a Japanese Statutory Corpus

--

Toward Construction of a Legal Term Ontology

Makoto Nakamura,

Yasuhiro Ogawa and Katsuhiko Toyama

(Nagoya University)


First of all


There wasn’t an English version of Japanese
statutes until 2009.


Some companies have their own database as
the need arises...


Japanese Law Translation
Database (JLT)


Released
in
2009.


Advanced user interface


Problems


quantity


quality


Problems of JLT


Quantity


307/7,700 laws so far (<4%) (2009
-
2013)


Revision control
wrt

amendment laws


Quality


A number of translators


Disunity of translation


l
aw titles, legal terms, etc.


specific expressions


Example: Law Titles

一般社団法人及び一般財団法人に関する法律

(
ippan

shadan

houjin

oyobi

ippan

zaidan

houjin

ni

kansuru

houritsu
)


Act Concerning General Corporations and General Foundations


Act on General Associations and Foundations


Act on General Associations and Incorporated Foundations


Act on General Incorporated Association and General Incorporated
Foundation


Act on General Incorporated Associations and General Incorporated
Foundations


Insurance Business Act (Act No.105 of 1995)

Article 148 (
4
) The
provisions of Article 78 (Liability for Damages Pertaining to
the Actions of the Representative) of the
Act on General Incorporated
Associations and General Incorporated Foundations

shall apply mutatis
mutandis to an Entrusting
Company......

Example:

Variations of Translations

この法律は、公布の日から施行する。

(
kono

houritsu

wa

koufu

no hi
kara

shikou

suru
.
)



This Act shall be enforced from the date of promulgation.


This Act shall come into effect as from the date of promulgation.


This Act shall come into effect as of the date of promulgation;


This Act shall come into effect as of the day of its promulgation.


This Act shall come into effect on the day of promulgation.


This Act shall come into force as from the date of its promulgation.


This Act shall come into force as from the day of its promulgation.


This Act shall come into force as of the date of its promulgation.


This Act shall come into force as of the day of promulgation.


This Act shall come into force from the day of promulgation.


This Act shall enter into force on the day of its
promulgation.


Problem of Quality


Law titles in different ways for citation


A variety of translations

An identical expression should have an identical translation for
the sake of the conservation of meaning.

Need of a translation
d
ictionary

Translation tools


Terminology (dictionary)


unification of translation


4,482 entries at the latest version


Ontology


systematic translation


Translation Memory
(
Sekine

et al., 2012)


Statistical Machine Translation

(Ogawa et al., 2013)


Tomorrow!!

Purpose


to construct a legal term ontology


to provide legal terms and their explanations








Approach


automatic extraction


application of natural language processing

Problems in This Study

1.
What are
legal
terms?



2.
Difficulty of Japanese text processing in the
legal domain



3.
Absence of a statutory corpus


1. What Are Legal Terms?

Classification

1.
Law titles

2.
Legal definitions

3.
Words in definitions

4.
Important terms


frequently used content words


characteristic words

5.
All words in laws

Our target

ガス事業法(昭和二十九年法律第五十一号)

第二条


の法律において「
一般ガス事業
」とは、
一般の需要に応じ導管
によりガスを供給する事業
をいう。


Gas Business Act (Act No.51 of 1954)

Article 2
The term “
General Gas Utility Business
” as used in this Act shall
mean
the business of supplying gas via pipelines to meet general
demand
.

1. What Are Legal Terms?

Legal Definitions


are explicitly defined in a law.


c
onsist of tuples of a defined term and its
explanation.


2. Difficulty of

Japanese Legal Text Processing


NLP tools:
low performance


special terms


syntactic rules peculiar to the legal domain


outdated characters



Simple methods for surface pattern recognition:
sufficient


f
requent use of boilerplate expressions

3
. Compilation of a

Japanese Statutory Corpus


includes all of the Japanese statutes


consists of 9,915 acts that have been enacted
up to 2012 since promulgation of the new
constitution of Japan in 1947


includes OCR errors


doesn’t follow


error correction, or


consolidation by amendment laws

Solutions of the Problems

1.
What are
legal
terms?


We focus on legal definitions.


1.
Difficulty of Japanese text processing in the
legal domain

→ Simple methods work effectively.


2.
Absence of a statutory corpus

→ We compiled a large corpus
.

Extraction Method


Simple pattern match from a legal text

a)
for independent sentences for definition


a sentence for definition


itemization

b)
for expressions parenthesized in a sentence


an expression for definition


the region of its explanation

a) Example of Surface
Pattern Rules


Input text:

ガス事業法

第二条

この法律において「一般ガス事業」とは、一般の需要に応じ導管によ
りガスを供給する事業をいう。

10

この法律において「ガス事業」とは、一般ガス事業、簡易ガス事業、ガ
ス導管事業及び大口ガス事業をいう。


Gas Business Act

Article 2
-
1 The term “General Gas Utility Business” as used in this Act shall
mean the business of supplying gas via pipelines to meet general demand.

Article 2
-
10 The term “Gas Business” as used in this Act shall mean a General
Gas Utility Business, Community Gas Utility Business, Gas Pipeline Service
Business or Large
-
Volume Gas Business.


Pattern of Definitions (6 rules)

/

(.+)
」とは、
(.+)(
を、
|
をいい、
|
をいう。
|
といい、
|
という。
|
とする。
)/

/”(.+)” (as used in this Act)? (shall
mean|means
) (.+)./

a
) Example of Surface
Pattern Rules


Input text:

ガス事業法

第二条

この法律において「
一般ガス事業
」とは、
一般の需要に応じ導管によ
りガスを供給する事業
をいう。

10

この法律において「
ガス事業
」とは、
一般ガス事業、簡易ガス事業、ガ
ス導管事業及び大口ガス事業
をいう。


Gas Business Act

Article 2
-
1 The term “
General Gas Utility Business
” as used in this Act shall
mean
the business of supplying gas via pipelines to meet general demand
.

Article 2
-
10 The term “
Gas Business
” as used in this Act shall mean
a General
Gas Utility Business, Community Gas Utility Business, Gas Pipeline Service
Business or Large
-
Volume Gas Business
.


Pattern of Definitions (6 rules)

/

(.+)
」とは、
(.+)
(
を、
|
をいい、
|
をいう。
|
といい、
|
という。
|
とする。
)/

/”
(.+)
” (as used in this Act)? (shall
mean|means
)
(.+)
./

a) Example of Surface Pattern
Rules for Itemization


Input text:

介護保険法

第七条6

この法律において「
医療保険各法
」とは、
次に掲げる法律
をいう。


健康保険法


船員保険法

・・・


Long
-
Term Care Insurance Act

Article 7 (6) The term “
Medical Insurance Acts
” as used in this Act means

the
following acts:

(i)
Health Insurance Act

(ii)
Seaman’s Insurance Act


.....


Pattern of Definitions for Itemization (8 rules)

/(

|

|

)(

(
各号
|
いずれか
))?

(
掲げる
|
定める
)/

/the following/

b) Example of Surface Pattern Rules
for Parenthesized Definitions


Input text:

民事訴訟法

第三十条4

第一項又は前項の規定により原告又は被告となるべき者を選定し
た者(以下「
選定者
」という。)は、その選定を取り消し、又は
選定された当
事者
(以下「
選定当事者
」という。)を変更することができる。


Code of Civil Procedure

Article 30 (4)
Persons who have appointed a party to stand as a plaintiff or
defendant pursuant to the provisions of paragraph (1) or the preceding
paragraph

(hereafter referred to as “
appointers
”) may rescind the appointment
or change
the party thus appointed
(hereafter referred to as the “
appointed
party
”).


Pattern of Parenthesized Definitions

/

(.+?)

(.+)


(
いう。
|
総称する。
)

/

/((.+?) referred to as “
(.+)
”)/

b) Extraction of Explanations


Extracting an explanation corresponds to finding its
starting point.


Explanations are likely to start from the beginning
of the sentence.


We restrict our target to extraction of explanations
satisfying this condition.


Easy!

民事訴訟法

第三十条4

第一項又は前項の規定により原告又は被告となるべき者を選定
した者
(以下「
選定者
」という。)は、その選定を取り消し、又は
選定され
た当事者
(以下「
選定当事者
」という。)を変更することができる。





Difficult!

If there is a parenthesis for a definition within
120 letters
from the beginning
without a comma or period, its
explanation is from the beginning to the last letter
before
the parenthesis
.

Experiments


Extraction of legal terms and their explanations
from:


independent sentences for definition


parenthesized expressions in a sentence


The
statutory
corpus


3,131
/ 9,915 statutes


w/o all laws concerning amendment and repeal of
pre
-
existing laws



Evaluation: precision and recall


Evaluation Method


Legal terms

1)
as precise as possible

2)
as many as possible



Evaluation

1)
Precision



2)

Recall

A

B

C


Grey:

the search space

A:

system’s response


B:

actual legal terms


C:

correct answers

Statutory
Corpus

=

|C|

|A|

=

|C|

|B|

Experimental Result

Definitions

# of
Tokens

# of
Types

Precision

Recall

in provisions

5,250

3,799

0.980

0.980

in parenthesis

9,624

6,030

0.850

0.396

Multiple defined terms


Japanese

English

# of laws

in provisions

中小企業者

small and medium sized
enterprise operator

53

老年者

senile

22

in parenthesis

旧法

former act

178

施行日

effective date

122

Legal Terms Extracted from

the Electricity Business Act

Defined

terms

Explanatory

sentences

Electricity

Business

General Electricity Business
and
Wholesale Electricity Business

Electricity Utility

General Electricity Utility
and
Wholesale Electricity Utility

Wholesale

Electricity Business

Business supplying a
General Electricity Utility
with electricity to be used for its
General Electricity
Business
for which the
Electric Facilities
that satisfy the requirements specified by an Ordinance of the
Ministry of Economy, Trade and Industry are used

Wholesale Electricity Utility

Person who has obtained a license under paragraph 1 of the next Article to conduct a
Wholesale Electricity
Business

General Electricity Business

Business supplying electricity to meet general demand

General Electricity Utility

Person who has obtained a license under paragraph 1 of the next Article to conduct a
General Electricity
Business

Designated Electricity Utility

Electricity

Utility
designated by an Ordinance of the Minister of International Trade and Industry

General Electricity Business Company

Company for

General Electricity Utilities

Electric Facilities

Machines, apparatuses, dams, waterways, reservoirs, electric lines, and other facilities installed for the
purpose of generating, transforming, transmitting, distributing or using electricity (excluding, however,
those installed in ships, vehicles or aircrafts, and those specified by a Cabinet Order)

Electric Facilities for General Use

Electric Facilities
for receiving electricity from another person at a voltage lower than that specified by an
Ordinance of the Ministry of Economy, Trade and Industry and using the electricity thus received at the
same premises where it has been received (including Small Capacity
Generation Facilities
installed in the
same premises and electrically connected with such
Electric Facilities
), which are electrically connected
with
Electric Facilities
installed outside the premises only via the electric lines used for receiving electricity

Electric Facilities for Private Use

Electric Facilities
other than
Electric Facilities
used for an
Electricity Business
and
Electric Facilities for
General Use

A Semantic Network of
Inclusive relations

Electricity Business

Electricity Utility

Wholesale Electricity Business

Wholesale Electricity Utility

General Electricity Business

General Electricity Utility

Designated Electricity Utility

General Electricity Business Company

Electric Facilities

Electric Facilities for General Use

Electric Facilities for Private Use

# of nodes

# of edges

38,974

26,969

Conclusion and Future Work


Legal terms and their explanations


over 14,000 terms with high precision


low recall rate for the process of parenthesis


applicable for any language


Toward legal term ontology


inclusive relation among explanations


systematic translation


Future works


anaphora resolution




Thank you!!

Legal Terms Referring to

the Term “Gas Business”

Legal Term

The title of Act

ガス事業

Gas Business

Gas Business

Act (Act No.51 of 1954)

ガス工作物

Gas Facilities

Gas Business Act (Act

No.51 of 1954)

料金

(gas) rate

Local Tax Act (Act No.226 of 1950

特定ガス消費機器

Specified Gas Appliance

Act concerning Supervising Installation Work of
Specified Gas Appliance (Act No.33 of 1979)

公益事業者

public benefit service

operator

Special Measures Act on Preparation, etc. for
Common
-
Use Tunnel (Act No.81 of 1963)

地方公営企業

local public enterprise

Act on Labor Relations of Local Public Enterprises
(Act No.289 of 1952)

事業者

business operator

Enterprise Rationalization of Promotion Act (Act
No.5 of 1952)

Patterns of Definitions

--

Insert Statements in Parenthesis

Defined term

Explanatory

sentence

(a)

in parenthesis

before parenthesis

A Gas Utility Service Provider or a Wholesale Gas Business Supplier

(these persons shall hereinafter be referred to as a “
da猠rtilityI et挮
”)
may, ...

(b)

before parenthesis

in parenthesis

In the prisoner of war camp,
chaplains, etc.

(i.e.
chaplains, and
prisoners of war who are permitted to engage in works listed in item
(iv) of Article 64 pursuant to the provision of Article 69
. The same shall
apply in paragraph (3) of Article 84) may ...

Patterns of Definitions


An independent provision



An insert statement in parenthesis

Patterns of Definitions


An independent provision







An insert statement in parenthesis

ガス事業法(昭和二十九年法律第五十一号)

第二条


の法律において「
一般ガス事業
」とは、
一般の需要に応じ導管
によりガスを供給する事業
をいう。


Gas Business Act (Act No.51 of 1954)

Article 2
The term “
General Gas Utility Business
” as used in this Act shall
mean
the business of supplying gas via pipelines to meet general
demand
.

Patterns of Definitions


An independent provision



An insert statement in parenthesis

ガス事業法(昭和二十九年法律第五十一号)

第四十二条

ガス事業者又は卸供給事業者
(以下「
ガス事業者等
」とい
う。)
は、その事業のように供するため、...


Gas Business Act (Act No.51 of 1954)

Article 42
A Gas Utility Service Provider or a Wholesale Gas Business
Supplier
(these persons shall hereinafter be referred to as a

Gas Utility,
etc.

)
may, when it is necessary to install pipelines on or under a road,
bridge, ditch, river, embankment or other public land in order to use such
pipelines for the businesses, use them with permission from the
administrator thereof to the extent that such use does not impair their
usability.

a) Independent Sentence for
Definition


ガス事業法(昭和二十九年法律第五十一号)

第二条


の法律において「一般ガス事業」とは、一般の需要に応じ導管
によりガスを供給する事業をいう。


Gas Business Act (Act No.51 of 1954)

Article 2 The term “General Gas Utility Business” as used in this Act shall
mean the business of supplying gas via pipelines to meet general
demand.

a) Independent Sentence for
Definition


ガス事業法(昭和二十九年法律第五十一号)

第四十二条

ガス事業者又は卸供給事業者(以下「ガス事業者等」とい
う。)は、その事業のように供するため、道路、橋、みぞ、河川、堤防
その他公共の用に供せられる土地の地上又は地中に導管を設置する必要
があるときは、その効用を妨げない限度において、その管理者の許可を
受けて、これを使用することができる。


Gas Business Act (Act No.51 of 1954)

Article 42 A Gas Utility Service Provider or a Wholesale Gas Business
Supplier (these persons shall hereinafter be referred to as a
“Gas Utility,
etc.”
) may, when it is necessary to install pipelines on or under a road,
bridge, ditch, river, embankment or other public land in order to use such
pipelines for the businesses, use them with permission from the
administrator thereof to the extent that such use does not impair their
usability.

a) Independent Sentence for
Definition


武力攻撃事態における捕虜等の取り扱いに関する法律(平成十六年法律
第百十七号)

第四十二条

宗教要員等(宗教要員及び第六十九条の規定により第六十四
条第四号に掲げる業務に従事することを許された捕虜をいう。第八十四
条第三項において同じ。)は、捕虜収容所内に置いて,被収容者の行う
第四十条に規定する宗教上の行為を補助し,又は前条第一項に規定する
宗教上の儀式を行うことができる。


Act on Treatment of Prisoners of War and Other Detainees in Armed
Attack Situations (Act No.117 of 2004)

Article 42 In the prisoner of war camp,
chaplains, etc.

(i.e. chaplains,
and prisoners of war who are permitted to engage in works listed in item
(iv) of Article 64 pursuant to the provision of Article 69. The same shall
apply in paragraph (3) of Article 84) may assist the detainees in
performing religious acts prescribed in Article 40 and may perform the
religious ceremonies prescribed in paragraph (1) of the preceding Article.

a) Independent Sentence for
Definition


武力攻撃事態における捕虜等の取り扱いに関する法律(平成十六年法律
第百十七号)

第百八条



審査会は、資格認定審査請求をしたもの(以下「資格認定
審査請求人」という。)が前項の期間内に補正をしないときは、裁決を
もって、資格認定審査請求を却下することができる。ただし、その不適
法が軽微なものであるときは、この限りでない。


Act on Treatment of Prisoners of War and Other Detainees in Armed
Attack Situations (Act No.117 of 2004)

Article 108 (2) In the case that a person who has appealed for a review
on the recognition of internment status (hereinafter referred to as
“the
applicant of the appeal for review on the recognition of internment
status”
) fails to correct the defect within the period set forth in the
preceding paragraph, the Review Board may dismiss the appeal for
review on the recognition of internment status by determination; provided,
however, that this shall not apply when such defect is minor.