Oral Defense of Doctoral Dissertation Volgenau School of Engineering, George Mason University Rommel Novaes Carvalho Bachelor of Science, University of Brasília, Brazil, 2003 Master of Science, University of Brasília, Brazil, 2008 Probabilistic Ontology: Representation and Modeling Methodology Tuesday, June 28, 2011, 2:00pm -- 4:00pm Nguyen Engineering Building, Room 4705 Committee Kathryn Laskey, Chair Paulo Costa Kuo-Chu Chang David Schum Larry Kerschberg Fabio Cozman Abstract The past few years have witnessed an increasingly mature body of research on the Semantic Web (SW), with new standards being developed and more complex problems being addressed. As complexity increases in SW applications, so does the need for principled means to cope with uncertainty in SW applications. Several approaches addressing uncertainty representation and reasoning in the SW have emerged. Among these is Probabilistic Web Ontology Language (PR-OWL), which provides Web Ontology Language (OWL) constructs for representing Multi-Entity Bayesian Network (MEBN) theories. However, there are several important ways in which the initial version PR-OWL 1.0 fails to achieve full compatibility with OWL. Furthermore, although there is an emerging literature on ontology engineering, little guidance is available on the construction of probabilistic ontologies. This research proposes a new syntax and semantics, defined as PR-OWL 2.0, which improves compatibility between PR-OWL and OWL in two important respects. First, PR-OWL 2.0 follows the approach suggested by Poole et al. to formalizing the association between random variables from probabilistic theories with the individuals, classes and properties from ontological languages such as OWL. Second, PR-OWL 2.0 allows values of random variables to range over OWL datatypes. To address the lack of support for probabilistic ontology engineering, this research describes a new methodology for modeling probabilistic ontologies called Uncertainty Modeling Process for Semantic Technologies (UMP-ST). To better explain the methodology and to verify that it can be applied to different scenarios, this dissertation presents step-by-step constructions of two different probabilistic ontologies. One is used for identifying frauds in public procurements in Brazil and the other is used for identifying terrorist threats in the maritime domain. Both use cases demonstrate the advantages of PR-OWL 2.0 over its predecessor.

1. Probabilistic Ontology:
Representation and Modeling
Methodology
Rommel Novaes Carvalho
Dissertation Defense
PhD in Systems Engineering and Operations Research
George Mason University
06/28/2011
Monday, June 27, 2011

2. Agenda
Introduction
Problem Statement
Contributions
Representing Uncertainty in Semantic Technologies
1st Major Contribution: PR-OWL 2.0
2nd Major Contribution: Uncertainty Modeling Process
for Semantic Technologies (UMP-ST)
Conclusion
2
Monday, June 27, 2011

3. Introduction
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 3
Monday, June 27, 2011

4. Summary of Problems
1st major problem - Mapping/Types
Probabilistic web ontology language (PR-OWL) does not have a well-
defined and complete integration between the deterministic and
probabilistic parts of an ontology
2nd major problem - Methodology
Probabilistic languages for semantic technologies like PR-OWL lack a
methodology for guiding the construction of models
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 4
Monday, June 27, 2011

5. Summary of Contributions
For the 1st problem - Mapping/Types
Extended probabilistic web ontology language (PR-OWL)
Led the development of a proof of concept tool in collaboration with
UnB [105]*
For the 2nd problem - Methodology
Developed a methodology for modeling probabilistic
ontologies (POs)
Created two use cases using the proposed methodology
*These numbers refer to the references in my dissertation
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 5
Monday, June 27, 2011

6. What’s the problem?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

7. What’s the problem?
Public Notices - Data
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

8. What’s the problem?
Information Gathering
Public Notices - Data
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

9. What’s the problem?
Information Gathering DB - Information
Public Notices - Data
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

10. What’s the problem?
> 1 Bi info
> 5 Tri US$
Information Gathering DB - Information
Public Notices - Data
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

11. What’s the problem?
> 1 Bi info
> 5 Tri US$
Information Gathering DB - Information
Public Notices - Data
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

12. What’s the problem?
> 1 Bi info
> 5 Tri US$
Information Gathering DB - Information
Public Notices - Data
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

13. What’s the problem?
Information Gathering DB - Information
Public Notices - Data Design - UnBBayes
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

14. What’s the problem?
Information Gathering DB - Information
Public Notices - Data Design - UnBBayes
Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

15. What’s the problem?
Information Gathering DB - Information
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

16. What’s the problem?
Information Gathering DB - Information
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 6
Monday, June 27, 2011

17. Semantic Web
Semantic Web (SW) is a web of data that can be
processed by machines [45]
E.g., allow machines to differentiate between 3 pounds (price of
product) and 3 pounds (weight of product)
Change focus from data driven to knowledge driven
The World Wide Web Consortium (W3C) states that
ontologies provide the cement for building the SW [46]
Ontology: Taken from Philosophy, where it means a systematic
explanation of being
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 7
Monday, June 27, 2011

18. Ontology
An ontology is an explicit, formal knowledge representation that
expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain;
Properties of those entities;
Relationships among entities;
Processes and events that happen with those entities;
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
The Web Ontology Language (OWL)
Developed by the W3C
As a language to represent ontologies for the SW
Accepted as a W3C recommendation in 2004
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 8
Monday, June 27, 2011

19. Ontology
An ontology is an explicit, formal knowledge representation that
expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities;
Relationships among entities;
Processes and events that happen with those entities;
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
The Web Ontology Language (OWL)
Developed by the W3C
As a language to represent ontologies for the SW
Accepted as a W3C recommendation in 2004
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 8
Monday, June 27, 2011

20. Ontology
An ontology is an explicit, formal knowledge representation that
expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities; firstName, lastName, procurementNumber, ...
Relationships among entities;
Processes and events that happen with those entities;
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
The Web Ontology Language (OWL)
Developed by the W3C
As a language to represent ontologies for the SW
Accepted as a W3C recommendation in 2004
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 8
Monday, June 27, 2011

21. Ontology
An ontology is an explicit, formal knowledge representation that
expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities; firstName, lastName, procurementNumber, ...
Relationships among entities; motherOf, ownerOf, isFrontFor ...
Processes and events that happen with those entities;
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
The Web Ontology Language (OWL)
Developed by the W3C
As a language to represent ontologies for the SW
Accepted as a W3C recommendation in 2004
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 8
Monday, June 27, 2011

22. Ontology
An ontology is an explicit, formal knowledge representation that
expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities; firstName, lastName, procurementNumber, ...
Relationships among entities; motherOf, ownerOf, isFrontFor ...
analyzing if requirements
Processes and events that happen with those entities; are met,
choosing best proposal, ...
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
The Web Ontology Language (OWL)
Developed by the W3C
As a language to represent ontologies for the SW
Accepted as a W3C recommendation in 2004
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 8
Monday, June 27, 2011

23. Ontology in OWL
9
Monday, June 27, 2011

24. Uncertainty in the SW
The community recognizes the need to represent
and reason with uncertainty
W3C created the URW3-XG in 2007
Concluded that standardized representations
were needed [50]
Probabilistic Web Ontology Language (PR-OWL)
is a candidate language to represent probabilistic
ontologies
Based on Multi-Entity Bayesian Network
(MEBN) logic
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 10
Monday, June 27, 2011

25. Probabilistic Ontology
A probabilistic ontology is an explicit, formal knowledge representation
that expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities; firstName, lastName, procurementNumber, ...
Relationships among entities; motherOf, ownerOf, isFrontFor ...
analyzing if requirements
Processes and events that happen with those entities; are met,
choosing better proposal, ...
Statistical regularities that characterize the domain;
Inconclusive, ambiguous, incomplete, unreliable, and dissonant knowledge related to entities
of the domain;
Uncertainty about all the above forms of knowledge;
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 11
Monday, June 27, 2011

26. Probabilistic Ontology
A probabilistic ontology is an explicit, formal knowledge representation
that expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities; firstName, lastName, procurementNumber, ...
Relationships among entities; motherOf, ownerOf, isFrontFor ...
analyzing if requirements
Processes and events that happen with those entities; are met,
choosing better proposal, ...
Statistical regularities that characterize the domain;
Inconclusive, ambiguous, incomplete, unreliable, and dissonant knowledge related to entities
of the domain;
Uncertainty about all the above forms of knowledge;
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 11
Monday, June 27, 2011

27. Probabilistic Ontology
A probabilistic ontology is an explicit, formal knowledge representation
that expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities; firstName, lastName, procurementNumber, ...
Relationships among entities; motherOf, ownerOf, isFrontFor ...
analyzing if requirements
Processes and events that happen with those entities; are met,
choosing better proposal, ...
Statistical regularities that characterize the domain;
Inconclusive, ambiguous, incomplete, unreliable, and dissonant knowledge related to entities
of the domain;
Uncertainty about all the above forms of knowledge;
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 11
Monday, June 27, 2011

28. Probabilistic Ontology
A probabilistic ontology is an explicit, formal knowledge representation
that expresses knowledge about a domain of application. This includes:
Types of entities that exist in the domain; Person, Procurement, Enterprise, ...
Properties of those entities; firstName, lastName, procurementNumber, ...
Relationships among entities; motherOf, ownerOf, isFrontFor ...
analyzing if requirements
Processes and events that happen with those entities; are met,
choosing better proposal, ...
Statistical regularities that characterize the domain;
Inconclusive, ambiguous, incomplete, unreliable, and dissonant knowledge related to entities
of the domain; P(isFrontFor|
valueOfProcurement = >1M,
Uncertainty about all the above forms of knowledge; annualIncome = <10k) = 90%
where the term entity refers to any concept (real or fictitious, concrete or abstract) that
can be described and reasoned about within the domain of application [2].
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 11
Monday, June 27, 2011

29. Probabilistic Ontology in PR-OWL 1.0
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 12
Monday, June 27, 2011

30. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

31. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

32. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

33. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

34. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

35. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

36. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

37. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

38. 1st Problem - Mapping/Types
?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

39. 1st Problem - Mapping/Types
?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

40. 1st Problem - Mapping/Types
?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 13
Monday, June 27, 2011

41. How to build Probabilistic Ontologies?
Information Gathering DB - Information
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

42. How to build Probabilistic Ontologies?
Information Gathering DB - Information
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

43. How to build Probabilistic Ontologies?
Information Gathering DB - Information
Logic
+
Uncertainty
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

44. How to build Probabilistic Ontologies?
Information Gathering DB - Information
?
Logic
+
Uncertainty
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

45. How to build Probabilistic Ontologies?
My objective is to define and
represent a context model for
the interoperability of Sensor
Networks. As my background is
not computer science, it's
being a little hard to
Information Gathering
understand how to put in DB - Information
?
practice a probabilistic
ontology.
PhD student, Wageningen University, The Netherlands Logic
+
Uncertainty
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

46. How to build Probabilistic Ontologies?
This seems a very promising
My objective is to define and tool, but we need to learn how
represent a context model for to best make use of it. When
the interoperability of Sensor we try to design using
Networks. As my background is UnBBayes, the questions we
not computer science, it's are trying to answer is how do
being a little hard to you identify which entities
Information Gathering
understand how to put in DB - Information
are relevant to the problem
?
practice a probabilistic and how translate them as
ontology. variables Logic system.
in your
PhD student, Wageningen University, The Netherlands Fusion Engineer, EADS Innovation Works, UK
+
Uncertainty
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

47. How to build Probabilistic Ontologies?
This seems a very promising
My objective is to define and tool, but we need to learn how
represent a context model for to best make use of it. When
the interoperability of Sensor we try to design using
Networks. As my background is UnBBayes, the questions we
not computer science, it's are trying to answer is how do
being a little hard to you identify which entities
Information Gathering
understand how to put in DB - Information
are relevant to the problem
?
practice a probabilistic and how translate them as
ontology. variables Logic system.
in your
PhD student, Wageningen University, The Netherlands Fusion Engineer, EADS Innovation Works, UK
I am evaluating PR-OWL as a +
knowledge representation as Uncertainty
well as reasoning formalism.
Public Notices I'd like to explore if/how it can
- Data Design - UnBBayes
be used for applications
using resource devices.
PhD student, University of Texas at Arlington, USA
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

48. How to build Probabilistic Ontologies?
This seems a very promising
My objective is to define and tool, but we need to learn how
represent a context model for to best make use of it. When
the interoperability of Sensor we try to design using
Networks. As my background is UnBBayes, the questions we
not computer science, it's are trying to answer is how do
being a little hard to you identify which entities
Information Gathering
understand how to put in DB - Information
are relevant to the problem
?
practice a probabilistic and how translate them as
ontology. variables Logic system.
in your
PhD student, Wageningen University, The Netherlands Fusion Engineer, EADS Innovation Works, UK
I am evaluating PR-OWL as a +
knowledge representation as Uncertainty these variables?
Why use
well as reasoning formalism. Why they are connected in
Public Notices I'd like to explore if/how it can
- Data such a way? How -do you
Design UnBBayes
be used for applications choose what type of
using resource devices. variable it is?
Fusion Engineer, EADS Innovation Works, UK
PhD student, University of Texas at Arlington, USA
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

49. How to build Probabilistic Ontologies?
This seems a very promising
My objective is to define and tool, but we need to learn how
represent a context model for to best make use of it. When
the interoperability of Sensor we try to design using
Networks.One thing which might be beyond the scope of the questions we
As my background is UnBBayes, this tutorial is a
not computer science, it'sof Modeling with MEBN". to answer is how do
write-up about "Art are trying Both narration and
being a little hard to
the resultant MEBN help in understanding the problem, but
you identify which entities
how one reach from a problem descriptionInformation at
Information Gathering
understand how to put in DB - to the problem
are relevantto a MEBN
?
practice a probabilistic clear. ... So when it how translate them as
times is not very and comes to MEBN, how
one decides about the context nodes, input nodes and system.
ontology. variables Logic resident
in your
PhD student, Wageningen University, The Netherlands
nodes? Most of the times it mightFusion Engineer, EADS Innovationbut UK
be pretty obvious Works,
I am evaluating PR-OWL as why certain nodes + modeled
sometimes it is not very cleara are
knowledge representation when they could also be modeled variables?
as input nodes in a fragmentas Uncertainty these
Why use
as context nodes, etc. Should we follow an object-oriented connected in
well as reasoning formalism. Why they are
I'd like to when identifying important entities or should we How -do you
Public Notices approach explore if/how it can
- Data such a way? Design UnBBayes
think used for applications logic, etc.? As a modeler what
be in terms of predicate choose what type of
using resource devices.
drives our thinking process? variable it is?
Fusion Engineer, EADS Innovation Works, UK
PhD student, University of Texas at Arlington, USA
Professor, Institute of Business Administration, Pakistan
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 14
Monday, June 27, 2011

50. 2nd Problem - Methodology
There is now substantial literature about
what PR-OWL is [2, 4, 5],
how to implement it [6-9], and
where it can be used [10-15]
There is an emerging literature on ontology engineering [4, 28]
But, little has been written about
how to model a probabilistic ontology
This lack of methodology is not only associated with PR-OWL
OntoBayes [30], BayesOWL [31], P-SHIF(D) and P-SHOIN(D) [32],
Markov Logic Network [33], Bayesian Logic [63], and Probabilistic
Relational Models [64], amongst others, do not have a methodology for
creating models
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 15
Monday, June 27, 2011

51. Contributions
For the 1st problem - Mapping/Types
Extended probabilistic web ontology language (PR-OWL) in
order to:
Provide a mapping between deterministic knowledge and probabilistic
knowledge
Allow reuse of existing types provided by OWL
Led the development of a proof of concept tool in
collaboration with UnB [105]
For the 2nd problem - Methodology
Developed a methodology for modeling probabilistic
ontologies (POs)
Created two use cases using the proposed methodology
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 16
Monday, June 27, 2011

52. Representing Uncertainty
in Semantic Technologies
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 17
Monday, June 27, 2011

53. Uncertainty in the SW
Deterministic SW will either consider a
statement to be true, false, or unknown
Shortcoming: no built-in support for uncertainty
In open world partial (not complete) or
approximate (not exact) information is more the
rule than the exception
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 18
Monday, June 27, 2011

54. Uncertainty in the SW
Deterministic SW will either consider a
statement to be true, false, or unknown
Shortcoming: no built-in support for uncertainty
In open world partial (not complete) or
approximate (not exact) information is more the
rule than the exception
∀x,y,z ((Mother(x,y) ∧ Mother(z,y)) Sibling(x,z))
∀x,y (Sibling(x,y) Related(x,y))
∀y∃x,z,r Committee(x,y)
∧ Participant(z,y) ∧ Responsible(r,z)
∧ Related(x,r) ViolationOfLaw(y)
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 18
Monday, June 27, 2011

55. Uncertainty in the SW
Deterministic SW will either consider a
statement to be true, false, or unknown
Shortcoming: no built-in support for uncertainty
In open world partial (not complete) or
approximate (not exact) information is more the
rule than the exception Committee(John,procurement34)
Participant(ITBusiness,procurement34)
∀x,y,z ((Mother(x,y) ∧ Mother(z,y)) Sibling(x,z)) Responsible(Richard,ITBusiness)
∀x,y (Sibling(x,y) Related(x,y))
∀y∃x,z,r Committee(x,y) Mother(John,Jane)
∧ Participant(z,y) ∧ Responsible(r,z) Mother(Richard,Jane)
∧ Related(x,r) ViolationOfLaw(y)
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 18
Monday, June 27, 2011

56. Uncertainty in the SW
Deterministic SW will either consider a
statement to be true, false, or unknown
Shortcoming: no built-in support for uncertainty
In open world partial (not complete) or
approximate (not exact) information is more the
rule than the exception Committee(John,procurement34)
Participant(ITBusiness,procurement34)
?
∀x,y,z ((Mother(x,y) ∧ Mother(z,y)) Sibling(x,z)) Responsible(Richard,ITBusiness)
∀x,y (Sibling(x,y) Related(x,y))
LivesAt(John,Address1)
∀y∃x,z,r Committee(x,y) LivesAt(Richard,Address1)
∧ Participant(z,y) ∧ Responsible(r,z) LastName(John,White)
LastName(Richard,White)
∧ Related(x,r) ViolationOfLaw(y)
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 18
Monday, June 27, 2011

57. Uncertainty in the SW
The community recognizes the need to represent
and reason with uncertainty
W3C created the URW3-XG in 2007
Concluded that standardized representations
were needed [50]
PR-OWL is a candidate language to represent
probabilistic ontologies
Based on MEBN logic
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 19
Monday, June 27, 2011

58. MEBN
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 20
Monday, June 27, 2011

59. MEBN
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 20
Monday, June 27, 2011

60. Why MEBN?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 21
Monday, June 27, 2011

61. Why MEBN?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 21
Monday, June 27, 2011

62. Why MEBN?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 21
Monday, June 27, 2011

63. Why MEBN?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 21
Monday, June 27, 2011

64. Why MEBN?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 21
Monday, June 27, 2011

65. PR-OWL 1.0
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 22
Monday, June 27, 2011

66. PR-OWL 1.0
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 22
Monday, June 27, 2011

67. UnBBayes - MEBN / PR-OWL 1.0
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 23
Monday, June 27, 2011

68. 1st Major Contribution
PR-OWL 2.0
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 24
Monday, June 27, 2011

69. 1st Problem - Mapping/Types
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 25
Monday, June 27, 2011

70. 1st Problem - Mapping/Types
?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 25
Monday, June 27, 2011

71. 1st Problem - Mapping/Types
?
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 25
Monday, June 27, 2011

72. Mapping Schema
PR-OWL
OWL
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

73. Mapping Schema
PR-OWL
OWL
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

74. Mapping Schema
PR-OWL
OWL
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

75. Mapping Schema
PR-OWL
OWL
hasEducationLevel
Person
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

76. Mapping Schema
PR-OWL
OWL
hasEducationLevel
Person EducationLevel
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

77. Mapping Schema
PR-OWL
OWL
hasEducationLevel
Person EducationLevel
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

78. Mapping Schema
PR-OWL
hasEducationLevel_RV
OWL
hasEducationLevel
Person EducationLevel
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

79. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
OWL
hasEducationLevel
Person EducationLevel
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

80. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
OWL _MFrag.person
hasEducationLevel
Person EducationLevel
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

81. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
OWL _MFrag.person
hasEducationLevel aspiresEducationLevel
Person EducationLevel
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

82. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
OWL _MFrag.person
hasEducationLevel aspiresEducationLevel
domain
Person EducationLevel
hasEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

83. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
OWL _MFrag.person
hasEducationLevel aspiresEducationLevel
domain
range
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

84. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
OWL _MFrag.person
Person
hasEducationLevel
hasEducationLevel
? aspiresEducationLevel
domain
range
EducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

85. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
OWL _MFrag.person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

86. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasArgument
hasEducationLevel_RV_person hasEducationLevel_RV_advisor
OWL _MFrag.person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

87. Mapping Schema
PR-OWL
hasEducationLevel_RV
“2”^^int
hasArgument
hasArgNumber
hasEducationLevel_RV_person hasEducationLevel_RV_advisor
OWL _MFrag.person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

88. Mapping Schema
PR-OWL
hasEducationLevel_RV
“2”^^int
hasArgument
hasArgNumber
hasEducationLevel_RV_person hasEducationLevel_RV_advisor
hasArgTerm
_MFrag.advisor
OWL _MFrag.person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

89. Mapping Schema
PR-OWL
hasEducationLevel_RV
“2”^^int
hasArgument
hasArgNumber
hasEducationLevel_RV_person hasEducationLevel_RV_advisor
hasArgTerm
_MFrag.advisor
OWL _MFrag.person
isSubsBy
Person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

90. Mapping Schema
PR-OWL
hasEducationLevel_RV
“2”^^int
hasArgument
hasArgNumber
hasEducationLevel_RV_person hasEducationLevel_RV_advisor
hasArgTerm
_MFrag.advisor hasEducationLevelAdvisor
OWL _MFrag.person
isSubsBy range
domain
Person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

91. Mapping Schema
PR-OWL
hasEducationLevel_RV
“2”^^int
?
hasArgument
hasArgNumber
hasEducationLevel_RV_person hasEducationLevel_RV_advisor
hasArgTerm
_MFrag.advisor hasEducationLevelAdvisor
OWL _MFrag.person
isSubsBy range
domain
Person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

92. Mapping Schema
PR-OWL
hasEducationLevel_RV
“2”^^int
hasArgument
hasArgNumber
hasEducationLevel_RV_person hasEducationLevel_RV_advisor
hasArgTerm isObjectIn
isSubjectIn
_MFrag.advisor hasEducationLevelAdvisor
OWL _MFrag.person
isSubsBy range
domain
Person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

93. Mapping Schema
PR-OWL
hasEducationLevel_RV
hasEducationLevel_RV_person
isSubjectIn
OWL _MFrag.person
hasEducationLevel
Person EducationLevel
hasEducationLevel
aspiresEducationLevel
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 26
Monday, June 27, 2011

94. Using Existing Types
*reproduced with permission from [2] - extended version
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

95. Using Existing Types
*reproduced with permission from [2] - extended version
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

96. Using Existing Types
*reproduced with permission from [2] - extended version
Boolean
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

97. Using Existing Types
*reproduced with permission from [2] - extended version
Boolean
Nominal
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

98. Using Existing Types
*reproduced with permission from [2] - extended version
Boolean
Nominal
Class
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

99. Using Existing Types
*reproduced with permission from [2] - extended version
Boolean
Nominal
Class X
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

100. Using Existing Types
*reproduced with permission from [2] - extended version
Boolean
DataType
Nominal
Class X
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

101. Using Existing Types
*reproduced with permission from [2] - extended version
Entity
Boolean
DataType
Nominal
Class X
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 27
Monday, June 27, 2011

102. PR-OWL 2.0 - Proof of Concept
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 28
Monday, June 27, 2011

103. PR-OWL 2.0 - Proof of Concept
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 29
Monday, June 27, 2011

104. PR-OWL 2.0 - Proof of Concept
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 30
Monday, June 27, 2011

105. PR-OWL 2.0 - Proof of Concept
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 31
Monday, June 27, 2011

106. PR-OWL 2.0 - Proof of Concept
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 32
Monday, June 27, 2011

107. 2nd Major Contribution
Uncertainty Modeling Process
for Semantic Technologies
(UMP-ST)
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 33
Monday, June 27, 2011

108. How to build Probabilistic Ontologies?
Information Gathering DB - Information
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 34
Monday, June 27, 2011

109. How to build Probabilistic Ontologies?
Information Gathering DB - Information
?
Logic
+
Uncertainty
Public Notices - Data Design - UnBBayes
Report for Decision Makers Inference - Knowledge
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 34
Monday, June 27, 2011

110. 2nd Problem - Methodology
There is now substantial literature about
what PR-OWL is [2, 4, 5],
how to implement it [6-9], and
where it can be used [10-15]
There is an emerging literature on ontology engineering [4, 28]
But, little has been written about
how to model a probabilistic ontology
This lack of methodology is not only associated with PR-OWL
OntoBayes [30], BayesOWL [31], P-SHIF(D) and P-SHOIN(D) [32],
Markov Logic Network [33], Bayesian Logic [63], and Probabilistic
Relation Models [64], amongst others, do not have a methodology for
creating models
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 35
Monday, June 27, 2011

111. Methodology
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 36
Monday, June 27, 2011

112. Modeling Cycle - Procurement Fraud
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

113. Modeling Cycle - Procurement Fraud
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

114. Modeling Cycle - Procurement Fraud
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

115. Modeling Cycle - Procurement Fraud
Goal: Find suspicious procurements
Query: Is there any relation between the committee and
the enterprises that participated in the procurement?
Evidence: They are siblings
They live at the same address
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

116. Modeling Cycle - Procurement Fraud
Goal: Find suspicious procurements
Query: Is there any relation between the committee and
the enterprises that participated in the procurement?
Evidence: They are siblings
They live at the same address
Person
Procurement
Enterprise
ownerOf
participatesIn
livesAt
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

117. Modeling Cycle - Procurement Fraud
Goal: Find suspicious procurements
Query: Is there any relation between the committee and
the enterprises that participated in the procurement?
Evidence: They are siblings
They live at the same address
Person
Procurement
Enterprise
ownerOf
participatesIn
livesAt
If a member of the committee
lives at the same address as a
person responsible for a bidder in
the procurement, a relationship is
more likely to exist between the
committee and the enterprises,
which lowers competition.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

118. Modeling Cycle - Procurement Fraud
Goal: Find suspicious procurements
Query: Is there any relation between the committee and
the enterprises that participated in the procurement?
Evidence: They are siblings
They live at the same address
Person
Procurement
Enterprise
ownerOf
participatesIn
livesAt
If a member of the committee
lives at the same address as a
person responsible for a bidder in
the procurement, a relationship is
more likely to exist between the
committee and the enterprises,
which lowers competition.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

119. Modeling Cycle - Procurement Fraud
Goal: Find suspicious procurements
Query: Is there any relation between the committee and
the enterprises that participated in the procurement?
Evidence: They are siblings
They live at the same address
Person
Procurement
Enterprise
ownerOf
participatesIn
livesAt
If a member of the committee
lives at the same address as a
person responsible for a bidder in
the procurement, a relationship is
more likely to exist between the
committee and the enterprises,
which lowers competition.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

120. Modeling Cycle - Procurement Fraud
Goal: Find suspicious procurements
Query: Is there any relation between the committee and
the enterprises that participated in the procurement?
Evidence: They are siblings
They live at the same address
Person
Procurement
Enterprise
ownerOf
participatesIn
livesAt
If a member of the committee
lives at the same address as a
person responsible for a bidder in
the procurement, a relationship is
more likely to exist between the
committee and the enterprises,
which lowers competition.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

121. Modeling Cycle - Procurement Fraud
Goal: Find suspicious procurements
Query: Is there any relation between the committee and
the enterprises that participated in the procurement?
Evidence: They are siblings
They live at the same address
Person
Procurement
Enterprise
ownerOf
participatesIn
livesAt
If a member of the committee
lives at the same address as a
person responsible for a bidder in
the procurement, a relationship is
more likely to exist between the
committee and the enterprises,
which lowers competition.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 37
Monday, June 27, 2011

122. Evaluation - Procurement Fraud
Case-based evaluation (integration test / scenarios)
Scenario Hypothesis (H) Evidence (E) Expected Result Result
(a) isSuspiciousProcurement (a) Low probability that (a) P(H = true | E) =
(procurement) ...does not P(H = true | E) 2.35%
1 support
(b) isSuspiciousCommittee hypothesis... (b) Low probability that (b) P(H = true | E) =
(procurement) P(H = true | E) 2.33%
(a) isSuspiciousProcurement ...does and does (a) P(H = true | E) =
(procurement) not support (a) 10% < P(H = true | E) < 50% 20.82%
2 hypothesis...
(b) isSuspiciousCommittee ...conflicting (b) 10% < P(H = true | E) < 50% (b) P(H = true | E) =
(procurement) information... 28.95%
(a) isSuspiciousProcurement (a) P(H = true | E) =
(procurement) (a) P(H = true | E) > 50% 60.08%
...support
3 hypothesis...
(b) isSuspiciousCommittee (b) 10% < P(H = true | E) < 50% (b) P(H = true | E) =
(procurement) 28.95%
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 38
Monday, June 27, 2011

123. Modeling Cycle - MDA
GO FASTER ON THIS
SLIDE! JUST SAY THAT I
USED THE SAME
METHODOLOGY ON A
DIFFERENT DOMAIN.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 39
Monday, June 27, 2011

124. Modeling Cycle - MDA
GO FASTER ON THIS
SLIDE! JUST SAY THAT I
USED THE SAME
METHODOLOGY ON A
DIFFERENT DOMAIN.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 39
Monday, June 27, 2011

125. Modeling Cycle - MDA
GO FASTER ON THIS
SLIDE! JUST SAY THAT I
USED THE SAME
METHODOLOGY ON A
DIFFERENT DOMAIN.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 39
Monday, June 27, 2011

126. Modeling Cycle - MDA
Goal: Identify whether a ship is a ship of interest
Query: Does the ship have a terrorist crew member?
Evidence: Crew member related to any terrorist;
Crew member associated with terrorist
organization
GO FASTER ON THIS
SLIDE! JUST SAY THAT I
USED THE SAME
METHODOLOGY ON A
DIFFERENT DOMAIN.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 39
Monday, June 27, 2011

127. Modeling Cycle - MDA
Goal: Identify whether a ship is a ship of interest
Query: Does the ship have a terrorist crew member?
Evidence: Crew member related to any terrorist;
Crew member associated with terrorist
organization
Ship
Person
Organization
isTerroristPerson
GO FASTER ON THIS hasCrewMember
SLIDE! JUST SAY THAT I isRelatedTo
USED THE SAME
METHODOLOGY ON A
DIFFERENT DOMAIN.
Introduction - Representing Uncertainty in ST - PR-OWL 2.0 - UMP-ST - Conclusion 39
Monday, June 27, 2011