Conceptual Models as Ontological Contracts Giancarlo Guizzardi - - PowerPoint PPT Presentation

▶
conceptual models as ontological contracts
SMART_READER_LITE
LIVE PREVIEW

Conceptual Models as Ontological Contracts Giancarlo Guizzardi - - PowerPoint PPT Presentation

Dec 25, 2023 132 likes •1.24k views

Conceptual Models as Ontological Contracts Giancarlo Guizzardi CORE/UNIBZ, Italy (together with Nicola Guarino and the NEMO Group) By Tom Gould I s D Ham I s Bacon Sausage I s Real-World ( Ontological ) Semantics X Formal

slide-1
SLIDE 1

Conceptual Models as Ontological Contracts

Giancarlo Guizzardi CORE/UNIBZ, Italy (together with Nicola Guarino and the NEMO Group)

slide-2
SLIDE 2

By Tom Gould

slide-3
SLIDE 3

Ham Bacon Sausage D

I

s

I

s

I

s

slide-4
SLIDE 4

Real-World (Ontological) Semantics X Formal Semantics

slide-5
SLIDE 5
slide-6
SLIDE 6

PERSON LIVING PERSON DECEASED PERSON SURGEON DONOR DONEE TRANSPLANT

slide-7
SLIDE 7
slide-8
SLIDE 8

“data are fragments of a theory of the real world, and data processing juggles representations of these fragments

  • f theory...The issue is ontology,
  • r the question of what exists.”

(G.H. Mealy, Another Look at Data, 1967)

slide-9
SLIDE 9
slide-10
SLIDE 10 PERSON LIVING PERSON DECEASED PERSON SURGEON DONOR DONEE TRANSPLANT HUMAN BEING PERSON SURGEON TRANSPLANT MEDICAL CERTIFICATION BODY MEDICAL LICENSE
slide-11
SLIDE 11

Semantic Interoperability

relating different worldviews, i.e., different conceptualizations of reality

=

slide-12
SLIDE 12 PERSON LIVING PERSON DECEASED PERSON SURGEON DONOR DONEE TRANSPLANT HUMAN BEING PERSON SURGEON TRANSPLANT MEDICAL CERTIFICATION BODY MEDICAL LICENSE

?

slide-13
SLIDE 13 PERSON LIVING PERSON DECEASED PERSON SURGEON DONOR DONEE TRANSPLANT HUMAN BEING PERSON SURGEON TRANSPLANT MEDICAL CERTIFICATION BODY MEDICAL LICENSE

?

slide-14
SLIDE 14
slide-15
SLIDE 15

“…was the crashing of two aeroplanes into New York's twin towers in September 2001 one event or two?” “In most disaster insurance, “occurrence” is carefully defined…”

slide-16
SLIDE 16

“IT WAS a $3.5 billion question: was the crashing of two aeroplanes into New York's twin towers in September 2001 one “In most disaster insurance, “occurrence” is carefully defined…”

slide-17
SLIDE 17
slide-18
SLIDE 18

“The insurers in the first trial had signed a form with a much tighter definition of “occurrence”…the insurance companies' claim that they always defined “occurrence” precisely”

slide-19
SLIDE 19

There is no doubt about the brute

  • reality. The issue is interpreting

that part of reality according to a certain system of categories

1

slide-20
SLIDE 20

There are multiple views on reality that can conflict and unless we are fully aware of their distinctions, we cannot safely harmonize those views

2

slide-21
SLIDE 21

The is no experiment that can be done to settle these conflicts. It can only be resolve by conceptual clarification and meaning negotiation relying on a prioristic system of categories

3

slide-22
SLIDE 22

Ontology as a Calculus of Content

  • For that we need a a prioristic system of categories

and their ties addressing issues of Identity, Unity (Parts and Wholes), Individuation, Change, Classification and Taxonomic Structures, Dependence (Existential, Historical, Relational, Notional), Causality, Essential and Accidental Characterization

  • We need Formal Ontology and Ontological

Analysis

slide-23
SLIDE 23

Ontology-Driven Conceptual Modeling

A discipline aiming at developing ontology-based methodologies, computational tools and modeling languages for the area of Conceptual Modeling

slide-24
SLIDE 24

UFO (Unified Foundational Ontology)

  • Over the years, we have built a Philosophically and Cognitively

well-founded Ontology to contribute to the general goal of serving as a Foundation for Conceptual Modeling

  • This Ontology has been used to as a theory for addressing may

classical conceptual modeling constructs such as Object Types, Identity and Taxonomic Structures (CAISE 2004, CAISE 2007, CAISE 2012, Synthese 2015, ER 2018), Part-Whole Relations (CAISE 2007, CAISE 2009, FOIS2010, CAISE 2011), Intrinsic and Relational Properties (ER 2006, ER 2008, ER 2011, CAISE 2015, DKE 2015, ER 2018), Weak Entities, Attributes and Datatypes (ER 2006), Events (ER 2013, BPM 2016), Multi-Level Modeling and Powertypes (JOWO 2015, ER 2015, DKE 2017, ER 2018), etc…

slide-25
SLIDE 25
slide-26
SLIDE 26
slide-27
SLIDE 27
slide-28
SLIDE 28
slide-29
SLIDE 29
slide-30
SLIDE 30
slide-31
SLIDE 31
slide-32
SLIDE 32
slide-33
SLIDE 33

K1 K2 K3 K4 K5

Kinds

slide-34
SLIDE 34

K1 K2 K3 K4 K5

Anti-Rigid Sortals (Roles and Phases)

slide-35
SLIDE 35

K1 K2 K3 K4 K5

Anti-Rigid Sortals (Roles and Phases)

slide-36
SLIDE 36

K1 K2 K3 K4 K5

Rigid Mixins

slide-37
SLIDE 37

K1 K2 K3 K4 K5

Anti-Rigid Mixins

slide-38
SLIDE 38

K1 K2 K3 K4 K5

Anti-Rigid Mixins

slide-39
SLIDE 39

Type Sortal Type MIXIN (e.g., insurable entity, cultural heritage item) Rigid Sortal Type

  • r KIND

(e.g., person, dog, organization car) Anti-Rigid Sortal Type including ROLES (e.g., student, singer) and PHASES (e.g., living person, metropolis)

slide-40
SLIDE 40
slide-41
SLIDE 41
slide-42
SLIDE 42
slide-43
SLIDE 43
slide-44
SLIDE 44

Person Man Adult Man British Citizen Singer Economist Young Boy Living Person

slide-45
SLIDE 45

Solution

  • 1. Characterizing the difference between:
  • NATURAL TYPE/KIND (e.g., PERSON) = RIGID SORTAL
  • ROLE (e.g., SINGER, ECONOMIST, BRITISH CITIZEN,

KNIGHT OF THE BRITISH EMPIRE) = ANTI-RIGID + RELATIONALLY DEPENDENT SORTAL

  • PHASE (e.g., LIVING PERSON, ADULT MAN) = ANTI-

RIGID + RELATIONALLY INDEPENDENT SORTAL

  • MIXIN (e.g., CULTURAL HERITAGE ENTITY, PHYSICAL

ENTITY, INSURABLE ITEM)? = MIXIN

slide-46
SLIDE 46

Role

  • All instances of a given ROLE are of the same KIND

(e.g., all Students are Person)

  • All instances of a ROLE instantiate that type only

contingently (e.g., no Student is necessarily a Student)

  • Instances of a KIND instantiate that ROLE when

participating in a certain RELATIONAL CONTEXT (e.g., instances of Person instantiate the Role Student when enrolled in na Educational Institution)

  • A ROLE cannot be a supertype of a Rigid Type ¡ ¡

6

slide-47
SLIDE 47

«kind» Person «role» Customer

slide-48
SLIDE 48

«role» Customer «kind» Person «kind» Organization

slide-49
SLIDE 49

WORLD W

Customer

Person

x

Instance of

slide-50
SLIDE 50

WORLD W

Customer

Person

x

Instance of Instance of

slide-51
SLIDE 51

WORLD W’

Customer

Person

x

Instance of Instance of

R-­‑

slide-52
SLIDE 52

WORLD W’

Customer

Person

x

Instance of Instance of

R-­‑ R+

slide-53
SLIDE 53

WORLD W’

Customer

Person

x

Instance of Instance of

R-­‑ R+

Instance of

slide-54
SLIDE 54

WORLD W’

Customer

Person

x

Instance of Instance of

R-­‑ R+

Instance of

We ¡run ¡into ¡a ¡logical ¡contradiction!

slide-55
SLIDE 55

Role

  • All instances of a given ROLE are of the same KIND

(e.g., all Students are Person)

  • All instances of a ROLE instantiate that type only

contingently (e.g., no Student is necessarily a Student)

  • Instances of a KIND instantiate that ROLE when

participating in a certain RELATIONAL CONTEXT (e.g., instances of Person instantiate the Role Student when enrolled in na Educational Institution)

  • A ROLE cannot be a supertype of a Rigid Type ¡ ¡

6

slide-56
SLIDE 56

«role»Student «kind» Person

NO! NO!

«role» Student «kind» Person «kind» Educational0Institution 0..n enrolled-at

«role» Student

NO!

slide-57
SLIDE 57
slide-58
SLIDE 58

The Emerging Role Pattern

«role» B «kind» A C enrolled+at

m..n ...

m ¡≥ 1

8

slide-59
SLIDE 59

The Emerging Phase Pattern

slide-60
SLIDE 60

Problem (2)

  • 1. Suppose that I want to represent that the ROLE

Customer can be played by entities of different KINDS, namely, People and Organizations. How to relate the ROLE and its allowed types using subtyping relations?

slide-61
SLIDE 61

A Classic Problem

«role» Customer «kind» Person «kind» Organization

slide-62
SLIDE 62

«role» Customer «kind» Person «kind» Organization

slide-63
SLIDE 63

«role»Customer Person Organization

A Possible Alternative?

slide-64
SLIDE 64

«roleMixin» Customer

slide-65
SLIDE 65

«roleMixin» Customer «role» PersonalCustomer «role» CorporateCustomer

slide-66
SLIDE 66

«roleMixin» Customer «role» PersonalCustomer Person Organization «role» CorporateCustomer

slide-67
SLIDE 67

The emerging RoleMixin Pattern

«roleMixin» A «role» B F D E «role» C 1..* 1..*

slide-68
SLIDE 68
slide-69
SLIDE 69

Conceptual ¡Modeling

Implementation1 ¡ Implementation2 ¡ Implementation3 ¡

DESIGN

slide-70
SLIDE 70

Conceptual ¡Modeling

Implementation1 ¡ Implementation2 ¡ Implementation3 ¡

DESIGN

by nemo

slide-71
SLIDE 71
slide-72
SLIDE 72

Complexity Management: Viewpoint Extraction, Modularization and Abstraction

slide-73
SLIDE 73

B A ¡ Valid ¡state ¡of ¡affairs ¡ according ¡to ¡the ¡representation Intended ¡state ¡of ¡affairs ¡ according ¡to ¡the ¡Conceptualization

slide-74
SLIDE 74

A

Under-­‑constraining

B Valid ¡state ¡of ¡ affairs ¡ according ¡to ¡ the ¡model Intended ¡state ¡of ¡affairs ¡ according ¡to ¡the ¡ ¡ Conceptualization

slide-75
SLIDE 75

Over-­‑constraining

B A Valid ¡state ¡of ¡ affairs ¡ according ¡to ¡ the ¡model Intended ¡state ¡of ¡affairs ¡ according ¡to ¡the ¡ Conceptualization

slide-76
SLIDE 76

B A B’ A’

False ¡Agreement

slide-77
SLIDE 77

Constraints

slide-78
SLIDE 78

Constraints

slide-79
SLIDE 79

Constraints

A ¡B

slide-80
SLIDE 80

Conceptual ¡Model ¡= ¡Structure ¡+ ¡Axiomatization ¡

A ¡B

slide-81
SLIDE 81

Conceptual ¡Model ¡= ¡Structure ¡+ ¡Axiomatization ¡ (Ontological ¡Commitment) ¡

A ¡B

slide-82
SLIDE 82

«role» Student «kind» Person «kind» Educational0Institution 1..* enrolled.at

□(∀x Person(x) → □(Person(x))) □(∀x Student(x) → ◊(¬Student(x))) □(∀x Student(x) → Person(x)) □(∀x Student(x) → ∃y Educational Institution(y) ∧ Enrolled-at(x,y)) …

9

slide-83
SLIDE 83

Conceptual ¡Model ¡= ¡ ¡ Structure ¡+ ¡Domain-­‑Independent ¡Axioms ¡+ ¡ ¡ Domain-­‑Specific ¡Axioms

B A

slide-84
SLIDE 84

ATL ¡Transformation Alloy Analyzer + OntoUML visual Plugin Simulation ¡and ¡Visualization

slide-85
SLIDE 85

«kind» Person «role»Organ Donor «role»Organ Donee «relator»Transplant «role» Transplant Surgeon 1 1..* «mediation» 1 1..* «mediation» 1..* 1..* «mediation»

slide-86
SLIDE 86
slide-87
SLIDE 87
slide-88
SLIDE 88
slide-89
SLIDE 89

Real-­‑Word ¡Semantics

slide-90
SLIDE 90

OntoUML Model Benchmark

  • Model benchmark with 56 models
  • Models in domains such as Provenance in Scientific

Workflow, Public Cloud Vulnerability, Software Configuration Management, Emergency Management, Services, IT Governance, Organizational Structures, Software Requirements, Heart Electrophisiology, Amazonian Biodiversity Management, Human Genome, Optical Transport Networks, Federal Government Organizational Structures, Normative Acts, and Ground Transportation Regulation

slide-91
SLIDE 91

The Emerging Anti-Pattern: Relation Between Overlapping Types (RelOver)

(a) AC, (b) IA and (c) RWOR.

slide-92
SLIDE 92

16

The Emerging Anti-Pattern: Relation Specialization (RelSpec)

slide-93
SLIDE 93

Heart ¡X Ventricle ¡Y Heart ¡Z Ventricle ¡W

17

slide-94
SLIDE 94

Heart ¡X Ventricle ¡Y Heart ¡Z Ventricle ¡W Heart ¡as ¡Pump ¡X Ventricle ¡as ¡Pump ¡W Heart ¡as ¡Pump ¡Z Ventricle ¡as ¡Pump ¡Y

18

slide-95
SLIDE 95

Anti-Pattern Catalogue

  • Association ¡Cycle ¡
  • Binary ¡Relation ¡Between ¡Over. ¡Types ¡
  • Deceiving ¡Intersection ¡
  • Free ¡Role ¡Specialization ¡
  • Imprecise ¡Abstraction ¡
  • Multiple ¡Relational ¡Dependency ¡
  • Part ¡Composing ¡Over. ¡Roles ¡
  • Whole ¡Composed ¡by ¡Over. ¡Parts ¡
  • Relator ¡Mediating ¡Over. ¡Types ¡
  • Relation ¡Composition ¡
  • Relator ¡Mediating ¡Rigid ¡Types ¡
  • Relation ¡Specialization ¡
  • Repeatable ¡Relator ¡Instances
  • Relationally ¡Dependent ¡Phase ¡
  • Generalization ¡Set ¡With ¡Mixed ¡Rigidity ¡
  • Heterogeneous ¡Collective ¡
  • Homogeneous ¡Functional ¡Complex ¡
  • Mixin ¡With ¡Same ¡Identity ¡
  • Mixin ¡With ¡Same ¡Rigidity ¡
  • Undefined ¡Formal ¡Association ¡
  • Undefined ¡Phase ¡Partition
slide-96
SLIDE 96
slide-97
SLIDE 97
slide-98
SLIDE 98
slide-99
SLIDE 99
slide-100
SLIDE 100

Anti-Pattern #Occ. #Error #Error / #Occ. #Refac. /#Error RelSpec 315 279 88.6% 97.1% RepRel 221 57 25.8% 84.2% RelOver 124 70 56.5% 77.1% BinOver 74 31 41.9% 74.2% AssCyc 20 14 70.0% 71.4% ImpAbs 125 11 8.8% 27.3% Total 879 462 52.56% 88.53%

slide-101
SLIDE 101
slide-102
SLIDE 102
slide-103
SLIDE 103
slide-104
SLIDE 104
slide-105
SLIDE 105
slide-106
SLIDE 106

“Few modelers, however, have had the experience of subjecting
 their models to continual, automatic review. Building a model incrementally with an analyzer, simulating and checking as you go along, is a very different experience from using pencil and paper alone. The first reaction tends to be amazement: modeling is much more fun when you get instant, visual feedback. Then the sense of humiliation sets in, as you discover that there’s almost nothing you can do right.” 
 
 (Daniel Jackson, Software Abstractions : Logic, Language, and Analysis, 2006)

slide-107
SLIDE 107

The Humble Modeler

[What] I have chosen to stress in this talk is the following. We shall do a much better modeling job in the future, provided that we approach the task with a full appreciation of its tremendous complexity,…,provided we respect the intrinsic limitations of the human mind and approach the task a Very Humble Modelers (paraphrasing Dijkstra’s Humble Programmer, 1972)

slide-108
SLIDE 108

For a primer into UFO and OntoUML…

  • GUIZZARDI, G., Ontological Patterns, Anti-Patterns

and Pattern Languages for Next-Generation Conceptual Modeling, 33rd International Conference on Conceptual Modeling (ER 2014), Atlanta, USA.

  • GUIZZARDI, G., WAGNER, G., ALMEIDA, J.P.A.,

GUIZZARDI, R.S.S., Towards Ontological Foundations for Conceptual Modeling: The Unified Foundational Ontology (UFO) Story, Applied Ontology, IOS Press, 2015.

slide-109
SLIDE 109

gguizzardi@unibz.it