Semiotics and conceptual modeling gv 2015
2 likes868 views
- Conceptual modeling in computer science often uses concepts that require interpretation, such as linguistic concepts, which challenges the model-theoretic semantics approach of formal logic. - Semiotics, as the study of signs and their interpretation, can provide a theoretical foundation for more formally and transparently addressing interpretation in conceptual models. - Ongoing research explores applying semiotic perspectives to linking ontologies and lexical resources to systematically represent the relationships between concepts, senses, and interpretations.
![Applicability of the Truth-schema
The problem of the definition of truth
obtains a precise meaning and can be
solved in a rigorous way only for those
languages whose structure has been
exactly specified.
At the present time the only languages
with a specified structure are the
formalized languages of various
systems of deductive logic.
[..] We are able, theoretically, to develop
in them various branches of science, for
instance, mathematics and theoretical
physics. [..] For other languages -- thus,
for all natural, "spoken" languages --
the meaning of the problem is more or
less vague, and its solution can have only
an approximate character.
Tarski, 1944
Many conceptual models are out
of the scope of the Truth-schema;
typically, those dealing with
linguistic concepts](https://image.slidesharecdn.com/semioticsandconceptualmodeling-gv2015-150526102529-lva1-app6891/85/Semiotics-and-conceptual-modeling-gv-2015-9-320.jpg)
![Semiotics
The snow
is white
WHITE SNOW
● Manifest significant entities have
no direct correspondence to extra-
linguistic entities
● Instead, they relate to mediating
entities, which in turn may relate to
extra-linguistic ones
● The resulting structure is called
sign
● Semiotics is an investigation about
sign relationships, their nature and
their interplay
Representamen
Expression
Signifier
Object
Referent
Interpretant
Content
Signified
sign
A sign [...] is something which stands to somebody for
something in some respect or capacity. The sign stands for
[...] its object [..] in reference to a sort of idea
C.S. Peirce, Collected Papers, 1897](https://image.slidesharecdn.com/semioticsandconceptualmodeling-gv2015-150526102529-lva1-app6891/85/Semiotics-and-conceptual-modeling-gv-2015-11-320.jpg)
Semiotics and conceptual modeling gv 2015
- 1. Conceptual Modeling in a Semiotic Perspective Guido Vetere IBM Italia, Center for Advanced Studies CNR, Istituto di Scienze e Tecnologie della Cognizione Centro ricerche interdisciplinare su cognizione, linguaggio e conoscenza dell’Università di Roma Tor Vergata 11 Maggio 2015 K Drive Knowledge Driven Data Exploitation FP7 grant 286348
- 2. Summary ● Attaining cognitive capabilities is one of the main trends of modern Computer Science (and industry) ● The research follows (and integrates) different approaches, based on evidence (data) and logic (theories) ● Logic-based approaches face the problem of providing symbols with some intepretation with respect to extra-logic entities ● However, formal logic at the basis of computer science is quite agnostic with respect to how such intepretation is given ● For every logic-based system in which interpretation is not trivial (e.g. social ones), this may result in a big issue ● However, addressing this issue is a relatively new concern (K. Liu, 2009, Semiotics in Information Systems Engineering) ● This talk is an introduction to the topic and a survey of some ongoing research
- 3. Conceptual Models ● Data Structures ● Database Schemas ● Industry Models ● Ontologies ● WordNets
- 4. The Logic Backbone ● Predicate First Order Logic (FOL) – Constants – Predicates – Variables – Connectives – Quantifiers ∀ x(B(x)→ A(x))∧(C (x)→ A(x))∧(B(x)→¬C (x))∧(C (x)→¬B(x)) A B C {disjoint}
- 5. The Logic Backbone ● Description Logic (FOL fragment) – Concepts – Roles – Individuals – Constructors – Assertions B⊆A,C⊆A, B∩C=∅ A B C {disjoint}
- 6. Logical Semantics ● Relation between expressions of a language and the objects (or states of affairs) referred to by those expressions ● A sentence (proposition) is true if and only if the corresponding state of affairs holds (Truth-schema) – “the snow is white” iff the snow is white Alfred Tarski, 1944 The Semantic Conception of Truth and the Foundations of Semantics WHITE (SNOW)
- 7. Logical Semantics ● Given – A logic language of individual constants, predicates, operators and inference rules – A theory, i.e. a set of valid formulas true by definition (axioms) – A model, i.e. a set of assignments of truth values to predicates with respect to individuals (interpretation), which fulfills the theory ● Infer the truth value of (well formed) logic formulas PERSON (JHON ), PERSON (MARY ) HATES(MARY , JHON ) Δ={JHON , MARY } Α={PERSON (), LOVES(,), HATES (,)} Λ=¬,∧, → ∀ x , y LOVES (x , y)→ PERSON ( X )∧PERSON (Y ) ∀ x , y HATES (x , y)→ PERSON ( X )∧PERSON (Y ) ∀ x , y HATES (x , y)→¬LOVES (x , y) LOVES (MARY , JHON )=F Alfred Tarski, 1944 The Semantic Conception of Truth and the Foundations of Semantics
- 8. Tarskian Semantics in Information Systems ● Software Programs – Runtime Memory = Model of data types structures ● Databases – Database Instance = Model of the Schema ● Semantic Web Linked Open Data – RDF Datasets = Model of some Ontology ● Knowledge Base (Graph) – Assertional Box = Model of the Ontology Activity={Patching ,Overlay ,Crack Sealing} interpretation
- 9. Applicability of the Truth-schema The problem of the definition of truth obtains a precise meaning and can be solved in a rigorous way only for those languages whose structure has been exactly specified. At the present time the only languages with a specified structure are the formalized languages of various systems of deductive logic. [..] We are able, theoretically, to develop in them various branches of science, for instance, mathematics and theoretical physics. [..] For other languages -- thus, for all natural, "spoken" languages -- the meaning of the problem is more or less vague, and its solution can have only an approximate character. Tarski, 1944 Many conceptual models are out of the scope of the Truth-schema; typically, those dealing with linguistic concepts
- 10. Semantics for Natural Languages ● Relativity – Different agents may supply different interpretations ● Vagueness – Many predicates can not be always clearly intepreted ● Creativity – Interpretations may be invented on the fly (and rapidly forgotten) The snow is white The bond between the signifier and the signified is arbitrary F. De Saussure, Cours de lingui- stique generale, 1916
- 11. Semiotics The snow is white WHITE SNOW ● Manifest significant entities have no direct correspondence to extra- linguistic entities ● Instead, they relate to mediating entities, which in turn may relate to extra-linguistic ones ● The resulting structure is called sign ● Semiotics is an investigation about sign relationships, their nature and their interplay Representamen Expression Signifier Object Referent Interpretant Content Signified sign A sign [...] is something which stands to somebody for something in some respect or capacity. The sign stands for [...] its object [..] in reference to a sort of idea C.S. Peirce, Collected Papers, 1897
- 12. Meaning Theories for Natural Language ● Correspondence – Aristotelism, logical positivism, L. Wittgenstein (Tractatus Logico-Philosophicus, 1921) – Speakers and listeners can verify truth conditions for sentences (T-scheme) – There’s a common access to a common World – Ontologies are given for everybody (realism) ● Interpretation – D. Davidson (Inquiries into Truth and Interpretation, 2001), H. Putnam (Mind, Language and Reality, 1975) – Listeners ascribe speakers consistent beliefs and honest communication intentions (principle of charity) – Listeners make hypotheses about speakers’ meaning intentions based on their own ontologies – Ontologies (conditions in the World) allow verifying interpretation hypotheses (externalism) ● Interplay – L. Wittgenstein (Philosophical Investigations, 1953), D.K. Lewis (Philosophical Papers I, 1983) – Listeners and speakers share linguistic rules by virtue of social exchanges (e.g. feedbacks) – Listeners understand speakers by making explicit reasoning on these rules – Ontologies are shared as long as they work within social linguistic environments (intersubjectivity, constructivism) ● Translation – W.V.O. Quine (Word and Object, 1960) – Speakers’ ontological commitments are not accessible by listeners – Listeners assign meanings to expressions on the basis of speakers’ observable behaviors – There are no shared ontologies (relativism) Reality Subjectivity
- 13. Types of Sign Signs can be studied from many perspectives
- 14. Types of Concepts N. Guarino et al, An Ontology of Meta-Level Categories, KR 94 ● Model-theoretic semantics: interpretation is not in question ● Still, it is possible to spot “interpretation- critical” areas Formal ontology focuses on different types of concepts
- 15. Vagueness Meta-Ontology P. Alexopoulos et al (2014), “A Metaontology for Annotating Ontology Entities with Vagueness Descriptions”, Springer. 2014. Vagueness is explicitely dealt with in recent proposals (FP7 K Drive Poject)
- 16. Linking Ontologies and Lexical Resources: the Semiotic Approach W3C Ontology-Lexicon Community Group, https://www.w3.org/community/ontolex/
- 17. Lexicon Ontology Interplay in Senso Comune If the sense S maps to the concept C, then there are entities of type C to which occurrences of S may refer to (ontological commitment) Non-Physical Entity Social Entity Sense Entity Information Object Physical Entity Endurant Substance water-1 commits-to (annotation) Expression noun-water has-sense G. Vetere, A. Oltramari, Lexicon Ontology Interplay in Senso Comune, LREC 2010
- 18. Conclusion ● Model-theoretic semantics of Logic and Formal Ontology delegates interpretation to “material” disciplines (e.g. Physics) ● Logic-based conceptual models in Computer Science make extensive use of concepts whose interpretation is in question (e.g. linguistic ones) ● As a result, interpretation is usually left to ad-hoc, opaque implementations ● Research is ongoing to provide more formal, transparent and systematic approaches ● Semiotics, as the “science of interpretation”, should be regarded to as the theoretical foundation of such development