=Paper=
{{Paper
|id=Vol-2518/paper-SOLEE1
|storemode=property
|title=Towards a Well-Founded Legal Domain Reference Ontology by Combining Conceptual Ontology Patterns
|pdfUrl=https://ceur-ws.org/Vol-2518/paper-SOLEE1.pdf
|volume=Vol-2518
|authors=Mirna El Ghosh,Habib Abdulrab
|dblpUrl=https://dblp.org/rec/conf/jowo/GhoshA19a
}}
==Towards a Well-Founded Legal Domain Reference Ontology by Combining Conceptual Ontology Patterns==
https://ceur-ws.org/Vol-2518/paper-SOLEE1.pdf
Towards a Well-Founded Legal Domain
Reference Ontology by Combining
Conceptual Ontology Patterns
Mirna EL GHOSHa,1 and Habib ABDULRABa
a
LITIS, INSA De Rouen, France
Abstract. Building well-founded domain ontologies is a great challenge in the
ontology engineering field. This concept has raised recently and it refers to
ontologies that are grounded in foundational ontologies. This paper addresses the
building of well-founded legal domain reference ontology by combining different
levels of conceptual ontology patterns. The ontology patterns are derived from the
foundational ontology UFO and the legal core ontology UFO-L. The use of the legal
domain ontology is demonstrated in the domain of carriage of goods by sea for
traceability purposes.
Keywords. Well-founded ontology, legal domain ontology, pattern-based ontology,
conceptual ontology patterns, UFO, UFO-L
1. Introduction
Building “well-founded” domain ontologies is a prominent challenge in the ontology
engineering field. This concept has been used mainly in Guizzardi’s works [1, 2] and it
refers to ontologies that are “grounded” in validated foundational ontologies. In other
words, concepts and relations in a well-founded domain ontology must be previously
analyzed in the light of a foundational ontology.
Generally, ontologies are classified according to their abstraction level into three
main categories [3]: foundational, core and domain. Foundational ontologies such as
UFO [1, 4, 5], that are located at the most abstract level, define a range of top-level
domain-independent ontological categories which form a general foundation for more
elaborated domain-specific ontologies. Underneath of foundational, core ontologies,
such as UFO-L [6] in the legal domain, are situated. They provide a precise definition of
structural knowledge in a specific field that spans across different domain applications.
At the lowest level, the domain ontologies, that describe the conceptualization related to
a specific domain (e.g. penal law, maritime law), are located. In addition, a relevant
classification of ontologies is proposed by Guizzardi [2] who differentiates between
reference and operational ontologies. Reference ontologies are particular kind of
conceptual models that are developed with the goal of making the best possible
description of the domain in reality [2]. Namely, when developing a reference ontology,
1
Corresponding Author. Copyright © 2019 for this paper by its authors. Use permitted under Creative
Commons License Attribution 4.0 International (CC BY 4.0).
the focus is on the expressivity of the representation and truthfulness to the domain being
represented [7]. Meanwhile, operational ontologies are machine readable
implementation version of reference ontologies [8]. Unlike reference ontologies,
operational ontologies are not focused on representation adequacy, but are designed with
the focus on guaranteeing desirable computational properties [9].
In the legal domain, building domain reference ontologies is a difficult task due to
the complexity of the domain and the difficulty of extracting semantic knowledge from
textual resources such as regulations and codes. It is important to notify that legal domain
ontologies differ from ontologies in other fields of practice, like medicine or engineering
in that they have to cover a wide range of common-sense concepts that are part of
physical, abstract, mental, and social worlds [10]. Legal domains share complex and
varied notions of norm and responsibility, but besides this, a legal domain refers to some
world of social activities [10].
The Unified Foundational Ontology (UFO) [5] is an example of a descriptive
foundational ontology that employ results from formal ontology, cognitive psychology,
linguistics and philosophical logics. In this context, reusing foundational and/or core
ontologies to support the development of domain ontologies is recognized as a promising
approach in the ontology engineering domain since it enables a speeding up of the
ontology development process [11]. Meanwhile, it is considered as a hard research issue
and one of the most challenging and neglected areas of ontology engineering [12]. The
problems of selecting the right ontologies to reuse, extending them and composing
several fragments have not been properly addressed yet [13].
Ontology patterns (OPs) are recognized as a promising approach to solve recurrent
ontology development problems [14]. OPs are modeling solutions that favor reuse of
encoded experiences and good practices [15]. In the ontology engineering community,
OPs have been addressed mainly in the works of [12,13,15, 16]. Recently, this approach
has gained more attention specially in [11,14,7] where its main goal is to support the
building of more consistent ontologies in a reuse-centered process. There are many
different types of OPs that can be used in different phases of the ontology engineering
process [11]. In this work, we are interested in Conceptual Ontology Patterns (COPs),
since the focus is on building a domain reference ontology in the legal domain.
The main goal of this paper is to build a well-founded legal domain reference
ontology by combining different levels of ontology patterns. The ontology patterns are
derived from the foundational ontology UFO [5] and the legal core ontology UFO-L [6].
After derivation, they will be combined to build the domain ontology. The targeted
ontology will be used for traceability in logistic networks in the domain of carriage of
goods by sea. The remainder of this paper is organized as follows: section 2 outlines the
unified foundational ontology UFO and the legal core ontology UFO-L. Section 3
describes the derivation of conceptual ontology patterns from UFO and UFO-L. In
section 4, the application of the ontology patterns is demonstrated in the domain of
carriage of goods for building a well-founded legal domain reference ontology. Finally,
section 5 outlines the related works and section 6 concludes the paper.
2. UFO and UFO-L
In this section, the unified foundational ontology (UFO) [1] and the legal core ontology
(UFO-L) [6] are introduced.
2.1. UFO
UFO [1] is a well-founded foundational ontology that employ results from formal
ontology, cognitive psychology, linguistics and philosophical logics. It makes a
fundamental distinction between Individuals and Universals. Individuals are entities that
exist in reality and obey a unique and determinate principle of identity, while Universals
are abstract patterns of features that can be realized in a number of different individuals
[7]. In UFO, two main kinds of individuals are distinguished: endurants and perdurants
[1]. Endurants are entities that are wholly present whenever they are present i.e. they
don’t have temporal parts [18]. They can be further specialized into Substantials
(Objects) and Moments (Tropes [17]). Substantials are existentially-independent
Endurants (e.g., a house, a person, the moon). Moments, or Tropes, in contrast, are
individuals that can only exist by inhering in other individuals [17]. Two main types of
moments are distinguished in UFO: Intrinsic moments and relators. Intrinsic moments
are moments that inhere in one single individual (e.g. the redness of a T-shirt). An
example of an intrinsic moment is a Mode (e.g. belief, intention, skill). Relators are
moments that depend on two or more endurants (e.g. marriage, enrollment).
Perdurants (events) are individuals composed of temporal parts and are existentially
dependent on endurants. They happen in time in the sense that they extend in time
accumulating temporal parts [19]. Examples of perdurants are a football game, a birthday
party or a business process. Therefore, two main layers of UFO are distinguished: the
layer A that consists of the ontology of substance and tropes individuals (UFO-A), the
layer B that consists of the ontology of events (UFO-B). In this paper, we are interested
in UFO-A, namely endurants and moments.
Concerning the Universals, mainly Endurant Universals (Figure 1), they are
composed of Substantial Universals and Moment Universals [1]. For the Substantials
Universals, UFO distinguishes between Sortal and Non-Sortal (Mixin) Universals [7].
Sortal universals are sortal types that either provide or carry a uniform principle of
identity for their instances [11,7]. Meanwhile, the Mixin universals, or Non-Sortals, are
universals that aggregate properties of distinct Sortals, i.e., it can have as instances
individuals obeying different principles of identity [7]. Within the category of Sortal
Universals, UFO differentiates between Rigid and Anti-Rigid sortals [11]. Kinds are
sortal rigid universals that provide a uniform principle of identity for their instances (e.g.,
Person). Subkinds are sortal rigid universals that carry the principle of identity supplied
by a unique Kind (e.g., a Kind Person can have the Subkinds Man and Woman that carry
the principle of identity provided by Person) [7]. Regarding Anti-Rigid, two main types
are identified: Role (e.g. Student) and Phase (e.g. Child). The meta-properties of rigidity
and anti-rigidity can be applied to Mixins where Rigid Mixins and Anti-Rigid Mixins are
distinguished. A Category (e.g. Physical Object aggregates essential properties of Table,
Car, Glass, etc.) represents a rigid mixin and a RoleMixin (e.g. Customer that aggregates
properties of Individual Customer and Corporate Customer) represents anti-rigid mixin.
In order to capture all these distinctions between endurants types, UFO-A has been
employed in the design of an ontologically well-founded conceptual modeling language
named OntoUML [5]. The modeling constructs in OntoUML are illustrated in the leave
categories in the hierarchy represented in Figure 1 [7]. Moreover, its metamodel contains
a number of formal constraints derived from the axiomatization of UFO that prescribe
the rules that govern the allowed combination of these constructs [7].
Figure 1. Fragment of UFO-A (Endurant Universals), adapted from [7].
2.2. UFO-L
UFO-L is a legal core ontology that uses domain-independent concepts provided by UFO
to represent essential concepts of law based on Alexy’s theory of fundamental rights
ontology [6]. UFO-L defines list of legal core concepts, such as Legal_Agent,
Legal_Object, Legal_Normative_Description, Legal_Moment, Legal_Norm,
Legal_Role, Legally_Defined_Event, Legal_Relator. By extending these concepts, a
conceptualization of the legal domain can be built. Moreover, UFO-L defines a reusable
modeling pattern (Legal_Relator pattern), illustrated in Figure 2. In this pattern, two
Legal_Agent play roles (Legal_Role) that are grouped into two different categories
(Legal_Role_Mixin). These categories are related through a Legal_Relator. The
Legal_Relator is grounded on an event relevant to the legal field called Legal_Event.
Legal_Relator mediates between legal categories of roles (Legal_Role) and consists of
legal moments (Legal_Moment) that are inherent in legal roles and externally dependent
on them. Legal moments are interlinked by correlation.
Figure 2. Legal_Relator pattern, adapted from [6].
3. Derivation of COPs from UFO and UFO-L
COPs are small fragments of ontology conceptual models that address a specific
modeling issue and can be directly reused by importing them in ontology under
development [12]. Thus, they are to be used during the ontology conceptual modeling
phase and focus only on conceptual aspects without any concern with the computational
part of the ontology [14]. COPs can be derived from either foundational ontologies
(Foundational Ontology Patterns - FOPs) or core/domain ontologies (Domain-Related
Ontology Patterns - DROPs) [11]. A COP extracted from a higher-level ontology can be
used to support the development of lower-level ontologies [7]. COPs should be encoded
in higher-order representation language [12] such as OntoUML [5]. This language has
been designed to reflect the ontological distinctions and axiomatization put forth by the
Unified Foundational Ontology (UFO) [4,5]. In the following, the conceptual ontology
patterns, FOPs extracted from UFO and DROPs extracted from UFO-L, are introduced
and illustrated in diagrams encoded in OntoUML [5].
3.1. Derivation of FOPs from UFO
Foundational ontology patterns (FOPs) are extracted from the foundations and rules of a
foundational ontology. A FOP is not a foundational ontology fragment; instead, it is a
self-contained set of related foundational rules and constraints that is applied to solve a
common modeling problem independently of domain [11]. Since FOPs are extracted
from foundational ontologies, they tend to be more generally applied and can be utilized
in isolation with weak dependencies with other patterns [8]. FOPs are reused by analogy
between the pattern and the problem in hand [14]. The result is an ontology fragment
with the FOP structure shaping the structures at the level of domain concepts [11]. FOPs
can be applied for building both core and domain ontologies. In this work, FOPs are
derived from the Unified Foundational Ontology (UFO). Three main FOPs patterns are
derived from UFO: Category, Role-Relator and Subkind.
3.1.1. Category Pattern
The Category pattern, depicted in Figure 3 as example, represents two main variants. In
the first variant (a), a kind generalization set collecting a disjoint set of kinds that
specialize the same Category. In the second variant (b), a simple Mixin specializing a
Category is illustrated.
Figure 3. Category FOP.
3.1.2. Role-Relator Pattern
The Role-Relator pattern, depicted in Figure 4, represents a Relator connected via
mediation relation to two different Roles that inherit the identity principle of exactly one
kind [11]. For instance, the relator marriage that connects two main roles Wife and
Husband that are inherited from Woman and Man respectively. This pattern is composed
of Role FOP where Role inherits Kind.
Role FOP
Figure 4. Role-Relator FOP.
3.1.3. Kind-Subkind Pattern
The Kind-Subkind pattern, depicted in Figure 5 as example, presents in two variants. In
the first variant (a), a simple subkind (or Kind) specializing a kind is illustrated. In the
second variant (b), a subkind (or Kind) generalization set collecting a disjoint set of
subkinds (or Kind) that specialize the same kind.
Figure 5. Subkind FOP.
3.2. Derivation of DROPs from UFO-L
Domain-related ontology patterns (DROPs) are reusable fragments extracted from core
or domain ontologies. They capture the core knowledge related to a given domain [14].
Therefore, DROPs are very inter-related and it is very difficult to apply them in isolation
[8]. DROPs, extracted from a core/domain ontology modeled already reusing FOPs, are
richer, carrying both structural and domain knowledge, characterizing a chained COP
application at the domain level [11]. DROPs are reused by extension, i.e. concepts and
relations of the pattern are specialized when the pattern is reused and also by including
new properties and relationships with the extended concepts [14]. Core ontologies are
important sources of DROPs, since they describe the core knowledge of a wide domain
that spans across different subdomains. Their models contain fragments of knowledge
that can be reused when modeling more specific domain ontologies [7]. For the
derivation of DROPs, the approach presented in [7] is applied. This approach is based
mainly on a fragmentation process that tends to extract sub-ontologies from UFO-L [6]
and splitting them into smaller pieces still meaningful to the domain. Therefore, two
main DROPs are extracted from UFO-L: Legal_Substance and Legal_Relator. This
process is guided by a list of Competency Questions (CQs) that can reveal modeling
needs in small pieces.
3.2.1. Legal_Substance Pattern
Legal_Substance pattern, depicted in Figure 6, represents the hierarchical structure of
Legal_Objects and Legal_Agents and their relationships. Two main CQs are addressed
for this pattern: (CQ1) How are Legal_Objects and Legal_Agents structured? (CQ2)
What categories are defined by Legal_Normative_Description?
Figure 6. Legal_Substance DROP.
3.2.2. Legal_Relator Pattern
UFO-L distinguishes two main types of Legal_Relator: Simple_Legal_Relator and
Complex_Legal_Relator. Right-Duty, NoRight-Permission, Power-Subjection and
Disability-Immunity are legal relators instantiating Simple_Legal_Relator.
Liberty_Relator instantiates Complex_Legal_Relator. In this section, Right-
Duty_Relator pattern, depicted in Figure 7, is extracted as DROP. Different CQs can be
addressed for this pattern such as [6]: (CQ1) Which agents are involved in the legal
relationship? (CQ2) What categories of legal roles are involved? (CQ3) What legal
moments compose the legal relationship? (CQ4) Who are the holders of each legal
moment? (CQ5) Whose legal moment is externally dependent? (CQ6) What event is the
basis of the legal relationship? (CQ7) is there a legal rule that defines the legal
relationship?
Figure 7. Right-Duty_Relator DROP.
4. Application of COPs for Building a Well-Founded Legal Domain Reference
Ontology
According to [11], modeling domain ontologies is not limited to the direct application of
patterns. Domain ontology fragments created from COPs are interrelated and need to be
put together. To do that, the ontology engineer can look for related DROPs, and also use
FOPs for combining the structure inherent to the different fragments [11]. Therefore,
when FOPs and DROPs are systematically applied in combination, the reuse is
maximized, the ontology building process become more productive, and the quality of
the resulting domain ontologies are improved [11]. In this section, reusing FOPs,
extracted from UFO, in combination with DROPs, extracted from UFO-L, is applied for
building a portion of a well-founded legal domain reference ontology in the domain of
carriage of goods by sea2 illustrated in Figure 8.
Figure 8. Building a Legal Domain Reference Ontology with different types of COPs.
For this portion of the legal domain reference ontology, that represents how carriage of
goods relationship is performed between carrier and shipper agents, seven competency
questions are defined: (CQ1) Which legal agents are involved in the carriage of goods
relator? (CQ2) What categories of legal roles are involved? (CQ3) What legal moments
compose the carriage of goods relationship? (CQ4) Who are the holders of each legal
moment? (CQ5) Whose legal moment is externally dependent? (CQ6) What legal event
2
Hague (The International Convention for the Unification of Certain Rules of Law relating to Bills of
Lading of August 1924) and Hamburg (The United Nations Convention on the Carriage of Goods by Sea 1978)
Conventions
is the basis of the carriage of goods relationship? (CQ7) is there a legal rule that defines
the carriage of goods relationship?
As noticed, the list of competency questions are specializations of the competency
questions of the legal core ontology UFO-L. Two main DROPs (Figures 6 and 7) are
applied to answer the list of the competency questions of the portion of the domain
ontology. In this context, the Right-Duty Relator DROP (Figure 7) inherits the structure
of the Role-Relator FOP (Figure 4) and the Legal_Substance DROP (Figure 6) inherits
the structure of the Category FOP (Figure 3). In addition, the Kind-Subkind FOP is
reused and applied by analogy in the domain ontology for representing two main
structures: (1) Shipper_Institution and Carrier_Institution are subkinds of
Agentive_Legal_Institution and (2) Contract_of_Carriage_of_Goods and
Bill_of_Lading are subkinds of Legal_Normative_Description. Therefore, FOPs and
DROPs are reused in combination for building a portion of the legal domain reference
ontology. Moreover, the combined reused has supported the axiomatization of the
targeted ontology by reusing and adapting the axioms defined for the COPs (FOPs and
DROPs) such as the disjointness axiom defined for Category (Figure 3) and Kind-
Subkind (Figure 5) patterns.
5. Related Works
The work presented in this paper is inspired mainly by the studies presented in [7,11]. In
previous works, such as [20], we have applied a reuse process of foundational and legal
core ontologies for building a well-founded legal domain ontology in the criminal
domain. The targeted ontology has been grounded in the foundational ontology UFO by
the application of the ontology-driven conceptual modeling language OntoUML. The
legal core ontology LKIF-Core [21] has been reused for representing the legal core
concepts and relations. We faced some difficulties during the ontology reuse process,
specifically on how to define the ontology parts to be reused and how the reuse process
will be applied. However, in this work, the application of ontology patterns is recognized
as a beneficial approach for building a well-founded legal domain ontology. Specifically,
by reusing patterns from the legal core ontology UFO-L which is modeled by reusing the
foundational ontology patterns of UFO. This strategy has led to a legal domain ontology
richer with the structural and domain knowledge.
6. Conclusion
This paper discussed the building of a well-founded legal domain reference ontology, in
the domain of carriage of goods by sea, by combining different levels of conceptual
ontology patterns (FOPs and DROPs). FOPs are derived from the foundational ontology
UFO and the DROPs are extracted from the legal core ontology UFO-L. The combined
reuse has enriched the domain ontology with structural and domain knowledge as well
as has contributed for reusing competency questions and axioms from foundational and
core ontologies [11]. Therefore, we can conclude that for building well-founded domain
ontologies it is essential and mandatory to reuse foundational and domain aspects in
combination by applying foundational and domain-related ontology patterns. Fore future
works, the targeted well-founded legal domain ontology will be used for building a
decision support system for the traceability in logistic networks.
Acknowledgements
This work has been supported by the European Union with the European Regional
Development Fund (ERDF) under Grant Agreement n°HN0002134 in the project
CLASSE 2 (“Les Corridors Logistiques: Application a la Vallée de la Seine et son
Environnement”).
References
[1] G. Guizzardi, PhD Thesis, 2005. Ontological Foundations for Structural Conceptual Models. Telematica-
Institut / CTIT.
[2] G. Guizzardi, 2007, On Ontology, ontologies, Conceptualizations, Modeling Languages and (Meta)Models,
in Proceedings of the 2007 conference on Databases and Information systems, 18-39.
[3] A. Scherp, C. Saathoff, T. Franz, S. Staab, Designing core ontologies. Applied Ontology 6 (2011), 177–
221.
[4] G. Guizzardi, G. Wagner, R. Guizzardi, Towards Ontological Foundations for Conceptual Modeling: The
Unified Foundational Ontology (UFO) Story, Applied Ontology 10 (2015), 259–271.
[5] G. Guizzardi, Ontological Foundations for Structural Conceptual Models, PhD Thesis, University of
Twente, 2005.
[6] C. Griffo, UFO-L, A Core Ontology of Legal Aspects Building Under the Perspective of Legal Relations,
PhD Thesis, Federal University of Espirito Santo, 2018.
[7] F. Ruy, G. Guizzardi, R. Falbo, C. Reginato, V. Santos, From Reference Ontologies to Ontology Patterns
and Back, Data and Knowledge Engineering 109 (2017), 41–69.
[8] Falbo, R. A., Barcellos, M.P., Nardi, J.C., Guizzardi, G. Organizing Ontology Design Patterns as Ontology
Pattern Languages, 10th Extended Semantic Web Conference, Montpellier, France, 2013.
[9] Falbo, R.D., Guizzardi, G., Gangemi, A., & Presutti, V. (2013). Ontology Patterns: Clarifying Concepts
and Terminology. WOP.
[10] V. Richard Benjamins, Pompeu Casanovas, Joost Breuker, Aldo Gangemi, Law and the Semantic Web:
Legal Ontologies, Methodologies, Legal Information Retrieval and Applications, 2005, Springer.
[11] F. Ruy, C. Reginato, V. Santos, R. Falbo, G. Guizzardi, Ontology Engineering by Combining Ontology
Patterns. In: Johannesson, P., Lee, M., Liddle, S., Opdahl, A., Pastor López, Ó. (eds.) ER 2015,
LNCS, 9381, pp. 173-186, Springer, Cham, 2015.
[12] A. Gangemi, V. Presutti, Ontology Design Patterns. In: Staab, S., Studer, R. (eds.) Handbook on
Ontologies, Springer, Heidelberg, 2009.
[13] E. Blomqvist, A. Ganemi, V. Presutti, Experiments on Pattern-based Ontology Design. In: K-CAP 2009,
pp. 41-48, ACM, USA, 2009.
[14] R. Falbo, G. Guizzardi, A. Gangemi, V. Presutti, Ontology Patterns: Clarifying Concepts and
Terminology. In: WOP 2013, vol. 1188, pp. 14-26, CEUR-WS, Germany, 2013.
[15] V. Presutti, E. Daga, A. Gangemi, E. Blomqvist, eXtreme Design with Content Ontology Design Patterns.
In: WOP 2009, pp. 83-97, ACM, USA, 2009.
[16] A. Gangemi, Ontology Design Patterns for Semantic Web Content. In: Gil, Y., Motta, E., Benjamins,
V.R., Musen, M.A (eds.) International Semantic Web Conference, LNCS, 3729, pp. 262-276, Springer,
Heidelberg, 2005.
[17] G. Guizzardi, G. Wagner, Towards an ontological foundation of discrete event simulation. In: Winter
Simulation Conference, pp. 652-664, 2010.
[18] G. Guizzardi, N. Guarino, J. Almeida, Ontological Considerations about the Representation of Events
and Endurants in Business Models. In: La Rosa, M., Loos, P., Pastor, O. (eds.) BPM 2016, LNCS, vol.
9850, pp. 20-36, Springer, Cham, 2016.
[19] G. Guizzardi, Towards Ontological Foundations for the Conceptual Modeling of Events. In: Ng, W.,
Storey, V.C., Trujillo, J.C. (eds.). ER 2013, LNCS, vol. 8217, pp. 327-341, Springer, Heidelberg, 2013.
[20] M. El Ghosh, PhD Thesis, 2018. Automation of Legal Reasoning and Decision Based on Ontologies.
Normandy University / INSA de Rouen.
[21] R. Hoekstra, J. Breuker, M. Di Bello and A. Boer, 2007, The LKIF Core ontology of basic legal concepts,
Proceedings of the Workshop on Legal Ontologies and Artificial Intelligence Techniques, 2007.