Suggesting Model Fragments for Sentences in Dutch Law

• Autore: Emile de Maat - Radboud Winkels
• Carica: Researcher at the 'Leibniz Center for Law' (LCL) - Associate Professor in Computer Science and Law at LCL of the Faculty of Law of the University of Amsterdam.
• Pagine: 185-202

Suggesting Model Fragments for Sentences in Dutch Law

EMIL E DE MA AT, RADBO UD WINKE LS∗

SUMM ARY:1. Introduction – 1.1. General Approach – 2. Sentences Dealing with the

Law – 2.1. Scope Declaration – 2.2. Repeal – 2.3. Insertion – 2.4. Replacement –

2.5. Renumbering – 2.6. Enactment Date – 2.7. Citation Title – 2.8. Application

Provision – 3. Sentences Dealing with Subject Matter – 3.1. Normative Sentences –

3.2. Def‌initions and Deeming Provisions – 4. Experiences – 5. Conclusions

1. INT RODU CTI ON

A main issue in the f‌ield of artif‌icial intelligence and law is the trans-

formation of sources of law that are written in natural language (and there-

fore rather informal) into formal models of law that computers can reason

with. This is a time and effort consuming process, error prone and different

knowledge engineers will arrive at different models for the same sources of

law. Moreover, these models should be closely linked to the original sources

(and at the right level of detail, i.e. isomorphic) since these sources tend

to change over time and maintenance of the models is a serious problem.

This calls for tools and a method for supporting this modelling process and

increasing inter-coder reliability.

We have been researching a method to create isomorphic models semi-

automatically, focusing on (Dutch) laws. This article presents a next step in

this creation process.

1.1. General Approach

In order to achieve (semi-)automatic modelling of sources of law, we fol-

low a number of steps, as shown in Fig. 1. The process starts with the source

document, written in natural language (Dutch). Currently, we focus on laws,

though we hope to expand to other types of sources of law later on. We f‌irst

make the structure of the document explicit, by marking up the different

parts, such as chapters, paragraphs and sentences, and assigning identif‌iers

to each part. We then proceed to mark all references within the source to

∗E. de Maat is Researcher at the “Leibniz Center forLaw” (LCL); R. Winkels is Associate

Professor in Computer Science and Law at LCL of the Faculty of Law of the University of

Amsterdam.

186 Informatica e diritto /Proceedings of the Workshop LOAIT 2010

other sources of law, using a parser based on patterns for references1. This

structure and reference information is stored in CEN/MetaLex XML2.

Fig. 1 – Steps in automatic modelling of legal texts

The next step is to create models for each individual statement in the

text. In most cases, each sentence in Dutch law forms a complete statement

(though possibly part of a bigger construct), so we are, in fact, creating a

model for each sentence in the text. In the last step, these individual models

are integrated with each other to come to a complete model. In order to cre-

ate the models, we start by classifying each sentence in the text as a specif‌ic

provision, such as a def‌inition, a duty, or a modif‌ication of an earlier law. In

total, we recognise ten different main categories. As with the references, this

is done by automatic recognition of certain patterns in the text3.

For several types of sentences, these patterns, together with some added

features, are suff‌icient to extract all information needed to create a model of

the sentence. This is usually the case with sentences that are about the law

itself, instead of the subject matter of the law. These sentences are discussed

in Section 2. Other sentences, such as obligations, do focus on the subject

matter, and can vary wildly. Simple patterns will not suff‌ice to deal with

these sentences, and to extract information from these types of sentences, we

1E. DE MAAT, R. WI NKEL S, T. VAN ENGER S,Automated Detection of Reference Struc-

tures in Law, in Engers T.M.van (ed.), “Legal Knowledge and Information Systems, Proceed-

ings of the Jurix 2006 Nineteenth Annual Conference”, Amsterdam, IOS Press, 2006, pp.

41-50.

2See http://www.metalex.eu/.

3E. DE MAAT, R. WI NKEL S,Automatic Classif‌ication of Sentences in Dutch Laws, in

Francesconi E., Sartor G., Tiscornia D. (eds.), “Legal Knowledge and Information Systems,

Proceedings of the Jurix 2008 Twenty-First Annual Conference”, Amsterdam, IOS Press,

2008, pp. 207-216.