A FRAMEWORK FOR SEMANTICALLY RICH LEGAL

DOCUMENTS AND APPLICATIONS

John Murphy and Robert Steele

Department of Computer Science, University of Technology Sydney, PO Box 123 Sydney, 2007, Australia

Keywords:

Semantic Web Application, Web Standards, Document Annotation, RDF, OWL.

Abstract:

This paper describes the use of Semantic Web technologies for enriching legal documents within a framework

that provides a base for the development of applications to reduce problems experienced by legal practitioners

in the administration of justice. The framework builds upon the terms and concepts found within structured

information sources in the document centric legal domain. Contributions of this paper include (1) the descrip-

tion of a bottom-up ontology development approach utilizing the terms and structures within legal documents;

(2) an explanation of how this legal knowledge can be captured from existing document resources and repre-

sented using Semantic Web languages; (3) an analysis of the comparative beneﬁts of this bottom-up approach

versus existing top-down approaches and (4) concrete examples of how to represent such legal knowledge

using OWL/ RDF. We argue such bottom-up legal ontology development grounded in the terms and concepts

found in existing legal resources will support direct and easier annotation of resources with the semantics of

legal concepts and hence provide a base for the easier development of a legal applications layer.

1 INTRODUCTION

Recent advances in semantic web technologies have

provided languages, tools and infrastructure that facil-

itate the application of these technologies to problems

in existing domains. The maturing of languages such

as OWL and RDF(S) with patterns of best practice has

enabled the application of Semantic Web technolo-

gies to domains that have the characteristics of be-

ing document centric with distributed, heterogeneous

and structured information sources. In this paper, we

report on an application of Semantic Web technol-

ogy to the legal domain and demonstrate the use of

these technologies to enrich the content of legal doc-

uments and reduce the problems faced by legal practi-

tioners dealing with complexissues using information

sources spanning multiple documents.

The domain of law shares similar characteristics

to the World Wide Web (WWW) because legal ad-

ministration is primarily document driven, with many

interrelated documents and independent authors. Le-

gal document publishing is characterized by differ-

ent types of related documents produced by differ-

ent organizations that have a mix of semi-structured

and free text parts. Unlike the WWW, the legal do-

main is underpinned by a formal legal vocabulary

that signiﬁcantly reduces the problems of ambiguity.

The characteristics of formal structure, reduced am-

biguity and its distributed heterogeneous nature make

the legal domain a strong candidate for the appli-

cation of Semantic Web technologies and standards

that encourage the integration of such information

resources. Web technologies have evolved to over-

come problems of distributed heterogeneous infor-

mation sources by standardizing access to informa-

tion resources. The Semantic Web technologies of

RDF(S) and OWL havefurther enhanced the ability to

process distributed information resources by enabling

the standardization of the description of a documents

structure and the meaning of its content.

Semantic Web technologies enable a standardized

and richer description of the meaning of legal docu-

ments by annotation of the documents terms in rela-

tion to an ontology. Traditional methods of describ-

ing the meaning of documents are achieved through

external categorization systems such as indexes and

digests, that assign meaning to the whole document

using tree like structures limited to expressing the sin-

gle is

kind Of relation between parent and child con-

cepts. OWL and RDF(S) are languages that enable the

formal deﬁnition of legal concepts and the relation-

ships between those concepts using logical constructs

193

Murphy J. and Steele R. (2008).

A FRAMEWORK FOR SEMANTICALLY RICH LEGAL DOCUMENTS AND APPLICATIONS.

In Proceedings of the Fourth International Conference on Web Information Systems and Technologies, pages 193-200

DOI: 10.5220/0001523601930200

 SciTePress

as well as providing the means to directly annotate

those concepts within a document. Further beneﬁts

of OWL are obtained by reasoning using the logi-

cal constructs from Description Logic that makes ex-

plicit the meaning that is implicit in the document. An

added beneﬁt of using web standards with public le-

gal documents is that it makes the information more

accessible to the wider population and provides a plat-

form that encourages broader participation in advanc-

ing research and application development in the legal

publishing domain. Many courts and legislators are

currently publishing directly on-line or sending their

documents to commercial legal publishers or organi-

zations such as AUSTLII (AustLII, 2007) for publi-

cation on the web.

Semantic Web technologies can be applied to

solve current restrictions on searching and under-

standing information. Searching legal documents can

be extended from “bag of words” or terms to search-

ing by concepts. The development of multiple prob-

lem speciﬁc ontologies enables different perspectives

of domain knowledge to be used by applications for

searching and navigation of a set of related documents

in a more focused and better targeted manner and pro-

vides the structure for understanding those documents

from a problem speciﬁc point of view.

This paper demonstrates the use of Semantic Web

technologies to deal with some of the problems faced

by legal practitioners in the legal domain. It describes

related semantic technology research in law, explains

where our research sits with current legal ontologyde-

velopment and describes the approach we have taken

to applying the Semantic Web technologies of OWL

and RDF to the enrichment of legal documents.

2 BACKGROUND

Ontology Development. A signiﬁcant part of legal

semantic research has been devoted to top-down on-

tology development to produce re-usable shared on-

tologies. The current top-down approaches to legal

ontology development have evolved from legal the-

ories such as those of (Kelsen, 1991), (Hart, 1961)

and (Bentham, 1970). These theories formed the ba-

sis of legal ontologies such as FOLaw (Valente, 1995)

that represent functional explanations of “how the law

works”. The next evolution in top-down legal ontol-

ogy development was the move toward representing

“what the law is” by capturing commonsense con-

cepts required for most applications of legal reason-

ing. LRI Core (Breuker and Hoekstra, 2004) is an

example of a core commonsense legal ontology that

reﬂects the roles, objects, functions and actions of the

legal system in society and are designed to facilitate

reasoning and argumentation.

Conceptual Search. Applicationsof semantic tech-

nology to conceptual searching are based on the idea

that established legal concepts can be modeled and

used for document retrieval systems that locate infor-

mation that conforms to a speciﬁc legal concept. Cur-

rent searching techniques often use syntactic or sta-

tistical analysis of terms rather than concepts or se-

mantic relations between terms. This often leads to

document retrieval that is not ranked by the expected

concept of relevance (van Noortwijk et al., 2005). For

systems to achieve the goal of conceptual retrieval

they require knowledge of legal concepts and issues

as well as knowledgeof the relationships of these con-

cepts to cases and legislation (Hafner, 1987). Our

framework establishes a base that applications can

build upon to help attain this goal.

Case Modeling. Modeling the relations between le-

gal principles is an important contribution to the task

of legal research in its own right, because some of

the associations between principals of law, legal is-

sues and facts in a sub-domain of law are not always

obvious to legal practitioners (Atkinson et al., 2005).

Recent developments in the modeling of legal argu-

mentation provide a structure for the development of

ontologies to speciﬁcally capture this important as-

pect of judicial process.

Legal Publishing on the Web. Public legal doc-

uments such as court judgements and legislation

are currently being published on websites such as

AUSTLII (AustLII, 2007). AUSTLII provides free

and open access to public legal documents on the

web using an index based on word occurrences and

databases of jurisdictions and parliaments for the pub-

lication of case law and legislation respectively. The

AUSTLII case law databases include hypertext links

to other cases, Acts of parliament and sections in

Acts.

In contrast to other frameworks that extract con-

cepts from the text using automated techniques (Bia-

gioli et al., 2005) and (Casellas et al., 2006), our work

concentrates on enriching existing legal documents

and the relationships between and within those docu-

ments focusing on those parts that provide procedural

context and summarization of the main legal points

as well as explicit external links that provide context

and supplementary information. Current approaches

to legal publishing on the web are generally restrained

to a document locator and optionally an anchor point

WEBIST 2008 - International Conference on Web Information Systems and Technologies

194

within the document. We argue there are signiﬁcant

beneﬁts to be gained in the ﬁelds of searching, docu-

ment retrieval, navigation and analysis of legal docu-

ments by enriching the documents structure, content

and links by enhancing or extending them with deeper

semantics. This we argue, will deliver more complex

and multi-faceted documents for analysis and shift the

development focus toward the concepts within docu-

ments rather than the document as a whole being the

fundamental level of conceptualization.

3 A FRAMEWORK FOR

SEMANTICALLY RICH LEGAL

DOCUMENTS AND

APPLICATIONS

The enrichment of legal documents through the ap-

plication of Semantic Web technologies begins with

the development of ontologies and the subsequent an-

notation of the existing legal documents. A bottom-

up approach to development is used taking advantage

of the structures and terminology found within legal

documents to build higher level legal concepts that

can be re-used in applications for document retrieval

and navigation, analysis and domain understanding,

document management, automated processing and le-

gal reasoning.

A good example of a structured legal document

is the common law Case Report that will be used

throughout this paper to give examples of our ontol-

ogy development approach, legal document annota-

tion process and the types of applications supported

by the framework. The Case Report document is an

important contributor to traditional paper-based legal

knowledge that contains the judgements of the court

as well as more formally structured parts contributed

by court reporters and editors as demonstrated in the

Case Report extract presented in ﬁgure 1.

Figure 1: Case Report Extract.

The extract is taken from the Head Note part of the

Case Report and shows a structured mix of judicial

case management information in the upper portion

and information concerning legal principles in the

portion marked “Held”. The Head Note in addition

to the court judgement, incorporates elements of ju-

dicial process such as hearing dates, parties and their

advocates, the presiding judge, the type of court, de-

scriptions of the nature of the case and cases cited.

Furthermore the Case Report includes more complex

parts such as the

“Catch Words”

block that is a hier-

archical categorization of key terms derived from Le-

gal Digests and a part that summarizes the important

legal issues dealt with by the case. Mixed with the

structured parts of the document are links to external

unifying sources such as common legal phrases, terms

from legal thesauri as well as links to relevant legisla-

tion.

The framework is composed of the three stages

of 1) ontology development, 2) document annotation

and 3) semantic application re-use. During the on-

tology development stage, higher level legal concepts

are evolved from the explicit legal information found

in the document text. The annotation stage involves

the semi-automated annotation of the legal text using

existing mark-up and knowledge of domain experts

such as Case Report editors whose current role in-

cludes the identiﬁcation and summarization of impor-

tant legal principles. The semantic application stage

involves the selection of relevant parts of the ontolo-

gies for re-use by speciﬁc applications for purposes

such as document retrieval and management and le-

gal reasoning.

3.1 Ontology Development

The framework adopts a bottom-up approach to on-

tology development using the structures and con-

cepts contained within legal documents as the start-

ing point. This approach to ontology development

begins with the identiﬁcation of concepts that are ex-

plicit in the legal documents including the names and

roles of the parties, the case citation, date of hearing,

the judges, the case and legislative citations, the ad-

vocates and references to the sections of any relevant

legislation. Figure 2 shows the conceptual model of

the Case Report document from which more abstract

concepts of law are develop to:

• represent the administrative rules and processes of

the court,

• encode the links between facts and legal princi-

ples,

• represent the roles of the participants in litigation,

A FRAMEWORK FOR SEMANTICALLY RICH LEGAL DOCUMENTS AND APPLICATIONS

195

• encode the links between the different types and

level of courts,

• encode the links between legal principles, judge-

ments and previous cases incorporating the con-

cepts of Following, Not Following, Afﬁrming, Ap-

plying and Reversing previous decisions and

• accommodate the representation of a sub-domain

of law using a structure that facilitates understand-

ing of the sub-domain. For example, the sub-

domain of negligence within the laws of “Tort”,

would require a structure that encompasses the el-

ements and sub-elements of negligence, “duty of

care”, “existence and breach of the duty of care”

and “damage suffered as a result of the breach”.

Figure 2: Case Report Concepts.

Bottom-up ontology development starts with the se-

lection and deﬁnition of of the most speciﬁc concepts

at the bottom of the hierarchy. The design advances

toward more abstract concepts by grouping the lower

concepts into more general concepts and by the ap-

plication of rules to the lower concepts for the con-

struction of higher level concepts. This bottom-up ap-

proach yields the beneﬁts of accessibility and famil-

iarity to the domain experts that annotate legal docu-

ments.

An example of developing higher level legal con-

cepts from speciﬁc concepts explicit in the document

is the Considered Case concept that is implied by the

citation link identiﬁed in the Case Report extract in

ﬁgure 1 as “AAA v BBB (1998) 23 FAM LR 716, af-

ﬁrmed”. This is the citation of a case that the judge

took into consideration when deciding the

KR v KR

case speciﬁcally Afﬁrming the previous decision. The

Considered Case is a higher level legal concept and

judicial mechanism that supports the related concept

of Case Authority in common law that encompasses

both the adoption or rejection of legal precedent and

deﬁned by the (NYSUC, 2007) as a:

previously adjudged action or decision on

same or similar point, serving as a rule or ex-

ample for present guidance.

Case Authority can be interpreted as the application of

cumulative weightings of importance to a case over

time using the decisions of subsequent judgements

that have considered the case and Followed, Not Fol-

lowed, Reversed or Applied the case rulings.

A number of beneﬁts can be obtained using the

framework that include:

• Accessibility by legal knowledge experts and le-

gal practitioners because the approach builds upon

the direct annotation of the original document

source and the legal domain is a knowledge in-

tensive domain that stores its knowledge in doc-

uments that can be marked-up using XML. This

direct access to the source by means of a readily

available technology gives non-technical person-

nel easier access to the source and enables broader

participation in the capture of knowledge from the

source documents.

• Familiarity of concepts used by domain experts

because top-down approaches require knowledge

of higher level concepts deﬁned by external par-

ties that are often highly abstract with subtle dis-

tinction that are not easy to understand and are

difﬁcult to use as a starting point for development.

• Better management of the development process

by encouraging partitioning of concepts along fa-

miliar domain and task speciﬁc lines.

Our research project is developing ontologies to

represent the knowledge of Case Report structures

and case reporting rules with a mixture of complex

and simple relations between legal concepts. The on-

tologies are designed to allow extension for the cap-

ture of complex legal principles and assist with tasks

such as indexing, extraction and summarization of le-

gal issues and the provision of an entry point for do-

main understanding and conceptual search for legal

practitioners.

3.2 Annotation

After the required legal concepts are deﬁned in on-

tologies, the lower level concepts are identiﬁed in the

documents and annotated. This process can be semi-

automated using existing mark-up to assist the iden-

tiﬁcation of terms. Lower level ontology concepts

can often be mapped directly to explicitly marked-

up terms in the text whilst other mark-up that iden-

tiﬁes document structures can be used to locate terms.

Our approach to the ontology development yields the

WEBIST 2008 - International Conference on Web Information Systems and Technologies

196

beneﬁt of starting with the most speciﬁc concepts that

are explicit within the legal document or can be easily

identiﬁed with human assistance.

A feature of legal documents that makes them

more conducive to a bottom-up approach is that much

of the publicly available legal documentation such

as Case Reports and legislation already have their

structure and many of the lower level explicit con-

cepts marked-up in XML or SGML. Many jurisdic-

tions have marked-up their court documents at the ini-

tial point of production by the the judge or judges as-

sociate and their work has been further expanded by

Case Reporters who add value to the judgements by

making explicit many concepts in the judgement doc-

ument that are implicit. A beneﬁt of the bottom-up

approach is that it only requires the mark-up of the

lower level legal concepts which facilitates the anno-

tation of this large body of existing legal documenta-

tion through automatic or semi-automatic transforma-

tions of the existing XML or SGML.

Further beneﬁts of direct annotation of legal doc-

uments are that it

• removes the impedance of external referencing of

sources found in other types of knowledge stores

such as databases or technologies acting external

to the source such as Xpointer.

• relieves the complications associated with au-

tomatic identiﬁcation of concepts using Natural

Language Processing by relying on the interven-

tion of human legal experts such as drafters of leg-

islation and Case Report editors.

A ﬁnal beneﬁt is the industry and research commu-

nity backing that is derived from the use of widely ac-

cepted standards based mark-up languages. This open

community backing ensures the availability of qual-

ity editing tools and repository and reasoning servers

along with a large body of published knowledgein the

Semantic Web technologies ﬁeld.

An example of the direct annotation of terms in a

legal document is shown in ﬁgure 3.

Figure 3 demonstrates the annotation of two of the

components of the example citation “AAA v BBB

(1998) 23 FAM LR 716, afﬁrmed” given previouslyin

the Case Report extract in ﬁgure 1. This is achieved

using the XML representation of OWL to mark-up

“AAA v BBB” as an instance of the Considered Case

concept that is implicit in the citation and to de-

ﬁne “AAA v BBB” as having a has considered value

property of “afﬁrmed” thereby deﬁning “AAA v

BBB” a previously considered case that has been af-

ﬁrmed.

Figure 3: Case Report Mark-up.

3.3 Semantic Applications

After the conceptual infrastructure of the ontologies

and annotated documents are created, the applica-

tions layer of the framework can be used to de-

velop applications for purposes including document

retrieval, navigation and domain understanding, doc-

ument management, legal process automation and le-

gal reasoning.

3.3.1 Document Retrieval, Navigation and

Domain Understanding

Understanding legal concepts and arguments is im-

portant in the legal domain and dealing in concepts

rather than documents enhances the access and un-

derstanding of legal knowledge particularly when the

understanding of a legal concept requires information

spanning multiple documents. In cases where docu-

ments are made available through web pages, HTML

provides simple links that have led to ﬁelds of knowl-

edge discovery built from the statistical analysis of

links into and out of a page, web site or domain. The

introduction of conceptual links brings greater com-

plexities to the link relationships and requires a Se-

mantic Web language to deﬁne and express the nec-

essary information. Links can be enriched using RDF

to hold more properties than the traditional resource

locator and it is possible for links to be multifaceted

and allow for reasoning over legal concepts by using

OWL to add higher level concepts. Potential bene-

ﬁts of this type of annotation are the development of

applications that bring about

• a reduction in time searching legal documents

• a reduction in the amount of non relevant material

the legal practitioner must process and

A FRAMEWORK FOR SEMANTICALLY RICH LEGAL DOCUMENTS AND APPLICATIONS

197

• a reduction of the amount of instruction a legal

practitioner must give to a computer to retrievethe

required information.

There is a view held that future search engines will

rely less on ranking of relevance of complete docu-

ments and behave more like information management

tools (Bontcheva et al., 2006, p143). These engines

will analyse and summarize the content of documents

rather than listing the documents by relevance and

leaving the determination of the documents meaning

to the user. Information will be deﬁned by meaning

and not statistical coincidence of terms and search-

ing will not be for documents, but for ideas or con-

cepts within documents or across documents. Legal

documents such as a Case Reports or legislation have

structures that deal with multiple concepts of law. The

evolution and reﬁnement of legal concepts across ju-

dicial processes such as court appeals and reapplica-

tion of the law to new facts, reinforces the importance

of understanding concepts across documents.

3.3.2 Document Management and Automated

Processing

A signiﬁcant problem in the legal publishing domain

is changes to documents such as legislation and its

contained law, and the changes to legal concepts as

case law evolves. This problem is separate to the sub-

stantive legal issues dealt with by legal practitioners

but is vital to ﬁnding and understanding the applica-

ble laws. The repeal of legislation, the afﬁrmation

or reversing of court decisions and the jurisdictional

boundaries and authority of courts are all examples

of metadata contained in legal documents that can be

enriched to provide better ways to manage legal doc-

uments to assist in better understanding the applica-

tion of the law. Document management of court cases

and legislation is particularly important for the main-

tenance of the state of the law because laws should

not be retrospectively applied by courts and therefore

it is essential to know the state of the law at the point

in time of the occurrence of some legal event.

Legislation has a number of stages that it must pass

before it becomes law. Courts also have a number

of steps that a document must pass through in or-

der to comply with court procedures. The annotation

of legal documents provides the information required

for systems to track the progress of the production of

these legal documents.

3.3.3 Legal Reasoning

Much effort has been invested by the AI community

to develop decision support systems built on differ-

ent forms of legal reasoning including, case based,

rule based and Description Logics (DL) based rea-

soning. Our approach limits the use of reasoning to

the evolution of higher level legal constructs. The

OWL language provides for reasoning using Descrip-

tion Logics and more recently, a constrained horn

clause language called the Semantic Web Rule Lan-

guage (SWRL). SWRL is a formal submission to the

W3C (Horrocks et al., 2004) for the introduction of

horn clause statements that can be used for the ex-

pression of rules such as those found in legislative

norms. The use of horn clauses to express legal norms

is demonstrated by (McCarty, 1989) in his Language

for Legal Discourse and also discussed in (Visser and

Bench-Capon, 1998) and can be used to express some

norms that can not be easily expressed using DL.

The legal concept of afﬁrmed on Appeal provides a

good example of the difﬁculty to assert in OWL DL

that a case Appealed and Afﬁrmed in the same hearing

is a case Afﬁrmed on Appeal. By contrast, the same

concept can easily be expressed in SWRL as

Af firmedOnAppeal(x) ← A f firmedIn(x, y) ∧

AppealedIn(x, z) ∧ sameHearingAs(y, z)

The reasoning capabilities of OWL DL can be used

to classify new concepts and can therefore be used

to create hypothetical scenarios that can be con-

structed using OWL class constructors and classiﬁed

for querying purposes.

A further beneﬁt of using the OWL language is that

the reasoning capabilities it provides eases the anno-

tation process in circumstances where a simple term

in the legal document is to be further developed as a

more complex legal concept. This beneﬁt is demon-

strated using the the previous case citation example

shown in the Case Report extract in ﬁgure 1. The

term Afﬁrmed is a member of a collection of attributes

for determining the impact a Considered Case has on

a previous court decision along with the terms Fol-

lowed, Not Followed, Applied and Reversed. At face

value, these terms are merely a list of discriminating

attributes for a Considered Case and would be created

as instances of an OWL class and be easily marked-

up in the legal document. If however the Afﬁrmed at-

tribute is to be expanded into a richer concept, it must

be created as an OWL class and therefore can not be

annotated directly in the document without also cre-

ating a speciﬁc

AAA v BBB

instance of the new Af-

ﬁrmed class.

Figure 4 demonstrates how the OWL language en-

ables the individual instance

AAA v BBB

to be de-

ﬁned as a type of anonymous class having some value

WEBIST 2008 - International Conference on Web Information Systems and Technologies

198

Figure 4: Case Citation Representation.

from Afﬁrmed Case for its has considered value

property.

This enables a reasoner to infer that

AAA v BBB

is an

Afﬁrmed Case

without having to create a par-

ticular instance of Afﬁrmed Case for

AAA v BBB’s

has considered value property. The term Afﬁrmed

can now be expanded as a legal concept to incorpo-

rate informationabout the jurisdictional hierarchy and

relative superiority of the court that heard the case or

could include a temporal dimension that indicates the

changing authority of a case over time. This exam-

ple demonstrates the use of new Semantic Web tech-

nologies and their associated patterns of best practice

(Rector and Welty, 2005) to enrich the concepts found

in legal documents in ways previously unavailable.

4 CONCLUSIONS

The Semantic Web technologies of RDF and OWL

enable existing structured and semi-structured legal

documents to be enriched to formally represent the

complex and interrelated concepts found within these

documents. We have described a framework that em-

ploys a bottom-up approach to developing legal on-

tologies grounded in the terms and structures of the

documents and re-using much of the existing mark-up

found in todays legal documents as well as the skills

of domain experts such as editors, case reporters and

legislative drafters. The framework also provides for

the direct annotation of these documents by the do-

main experts and the selective use of the the ontolo-

gies in semantic applications. We have also discussed

the beneﬁts of this approach and give concrete ex-

amples in OWL and RDF. It is considered that such

an approach will facilitate more direct and easier le-

gal document annotation, as the ontology concepts are

groundedin the existing terms found in the documents

themselves.

REFERENCES

Atkinson, K., Bench-Capon, T., and McBurney, P. (2005).

Arguing about cases as practical reasoning. In ICAIL

2005: Proceedings of the 10th international confer-

ence on Artiﬁcial intelligence and law, pages 35–44.

ACM Press, New York, NY, USA.

AustLII (2007). Australasian legal information institute.

http://www.austlii.edu.au/.

Bentham, J. (1970). An Introduction to the Principles of

Morals and Legislation. Athlone Press, London.

Biagioli, C., Francesconi, E., Passerini, A., Montemagni,

S., and Soria, C. (2005). Automatic semantics ex-

traction in law documents. In ICAIL 05: Proceedings

of the 10th international conference on Artiﬁcial in-

telligence and law, pages 133–140. ACM Press, New

York, NY, USA.

Bontcheva, K., Davies, J., Duke, A., Glover, T., Kings, N.,

and Thurlow, I. (2006). Semantic Web Technologies:

trends and research in ontology-based systems, chap-

ter 8, pages 139–169. John Wiley & Sons, Ltd.

Breuker, J. and Hoekstra, R. (2004). Epistemology and on-

tology in core ontologies: Folawand lri-core, two core

ontologies for law. In Proceedings of EKAW Workshop

on Core ontologies. CEUR.

Casellas, N., Jakulin, A., Vallb, J.-J., and Casanovas, P.

(2006). Acquiring an ontology from the text. In Ali,

M. and Dapoigny, R., editors, Advances in Applied Ar-

tiﬁcial Intelligence, pages 1000–1013. Springer.

Hafner, C. D. (1987). Conceptual organization of case law

knowledge bases. In ICAIL ’87: Proceedings of the

1st international conference on Artiﬁcial intelligence

and law, pages 35–42, New York, NY, USA. ACM

Press.

Hart, H. (1961). The Concept of Law. Clarendon Press,

Oxford.

Horrocks, I., Patel-Schneider, P. F., Boley, H., Tabet, S., and

Dean, B. G. M. (2004). SWRL: A Semantic Web Rule

Language Combining OWL and RuleML. Technical

report, W3C.

Kelsen, H. (1991). General Theory of Norms. Clarendon

Press, Oxford.

McCarty, L. T. (1989). A Language for Legal Discourse.

In Proceedings of the Second International Confer-

ence on Artiﬁcial Intelligence and Law, pages 180–

189, Vancouver, Canada.

NYSUC (2007). New York State Uniﬁed Court System,

law libraries, glossary of legal terms.

Rector, A. and Welty, C. (2005). Simple part-whole re-

lations in OWL Ontologies. Best practices, W3C,

http://www.w3.org/2001/sw/BestPractices/OEP/ Sim-

plePartWhole/. Retrieved August 2007.

A FRAMEWORK FOR SEMANTICALLY RICH LEGAL DOCUMENTS AND APPLICATIONS

199

Valente, A. (1995). Legal knowledge engineering: A mod-

elling approach, volume 30 of Frontiers in artiﬁcial

intelligence and applications. IOS Press.

van Noortwijk, K., Visser, J., and Mulder, R. V. D. (2005).

Re-usable retrieval concepts for the classiﬁcation of

legal documents. In Proceedings of the 10th inter-

national conference on Artiﬁcial intelligence and law,

pages 252–253. ACM Press, Bologna, Italy.

Visser, P. R. S. and Bench-Capon, T. J. M. (1998). A

comparison of four ontologies for the design of legal

knowledge systems. Artiﬁcial Intelligence and Law,

6(1):27–57.

WEBIST 2008 - International Conference on Web Information Systems and Technologies

200