Knowledge Interoperability

Ontological Browser (Biblioteca Palafoxiana, Puebla, Mexico)

Ontologies: A Brief Reminder

As the brainchild of classical philosophers ontologies are meant to provide a formal and systematic account of what makes sense in a given domain of discourse. For enterprises, and more generally for organizations, it implies a twofold integration of symbolic contents and semantics.

Contents Integration

Symbolic contents must be organized according to their status:

  • Terms, employed to name individuals entries whatever their nature (glossaries or lexicons)
  • Facts from environments (data)
  • Categories, meant to make sense of facts and define objectives, organization, and systems (models)
  • Documents, vessels used to store, exchange, or communicate contents
  • Concepts, for organizing cognitive (aka mental) representations independently of reifications and/or representations
Anatomy of Ontologies

Then, since contents are expressed through languages, they must be mapped to linguistic dimensions:

  • Syntactic: rules defining how the terms can be combined
  • Lexical: for the individual meaning of the terms employed
  • Semantic: for the meaning of the syntactic constructs (aka phrases)
  • Pragmatic: for the meaning of the semantic constructs depending on contexts

Finally, in order to ensure their interoperability, ontologies must maintain an epistemic distinction between what is known and how is it known.

That organization of contents in terms of representation (thesauruses, models, ontologies), epistemic nature (modalities), and language constitutes the backbone of ontologies.

Semantic Interoperability

Ontologies should ensure the consistency and interoperability of symbolic resources (data, information, knowledge) independently of contexts and concerns. The first objective is thus to avoid the ambiguous, contradictory, and circular definitions thwarting most institutional standards; to that end the kernel introduces a core of semantic axioms:

  • Instances: Physical, digital, or symbolic occurrences of objects or phenomena. Instances are represented by individuals in OWL
  • Attributes: characteristics of instances that belong to them and make them recognisable.
  • Collections: Set of instances managed uniformly, independently of their nature.
  • Identity: Unique value associated with an instance: External identities are defined independently of the organization or system under consideration
  • Behavior: ability of an instance to change the state of instances, including itself. Objects are either active (with behavior) or passive (without behavior), and the propriety is exclusive.
  • State: Named sets of values that characterize instances of objects (actual or symbolic), processes, or representations, between events.
  • Event: Change in the state of objects and phenomena. External events are changes triggered from outside the organization or system under consideration.

It must be stressed that these axioms are not meant to be universal: like music key they aim is only to provide a reliable and open basis for acyclic networks of definitions.

Axioms serve as roots and firebreaks in acyclic semantic networks

Besides OWL conceptual hierarchies (yellow color), the kernel uses semantic connectors (blue color) for standard (homonyms, synonyms, antonyms, …) and specific associations to ensuring the forward (thesaurusFw_) and backward (thesaurusBw_) traceability of definitions and eliminate circular references.

Ontological Prism

Facts, Concepts, Categories

The immersion of enterprises organization and systems in digital environments calls for a change of paradigm that could take into account the difference between:

  • Data, for facts as defined in environments
  • Information, for the categories used to manage symbolic representations
  • Knowledge, for the concepts used to define enterprises’ objectives, assets, and value chains

That can be achieved by a simple shift of perspective from layered pyramids to prisms ensuring a seamless and consistent integration of thesauruses, models, and knowledge graphs:

Ontological diffraction & the Fabric of Knowledge

That ontological diffraction open the door to interoperability at models and conceptual levels.

Logical & Functional Interoperability

Within enterprises semantics can be neatly set by business, organizational and system engineering contexts and concerns. But that’s not the case for the open-ended semantics of targeted environments. In order to integrate domain-specific semantics, the kernel uses a core of commonly accepted stereotypes based on the semantic axioms introduced above and set along the standard twofold taxonomy of Symbolic/Physical and Object/Behavior:

  • Physical/Digital environment: locations, persons, active and passive objects, events, and processes
  • Symbolic/Business environment: domains, organizational entities, business entities, roles, and activities

Ontological prisms can then be used to weave together symbolic and physical realms through observation (new facts), reasoning (new models), and judgment put to test (experience).

Ontological diffraction & the Engineering of Knowledge

Enterprise architecture provides a testbed of functional interoperability, enabling the alignment of EA governance issues with enterprises’ symbolic resources:

  • Requirements (facts)
  • Data analysis (facts/concepts)
  • Business analysis (concepts)
  • Strategic planning (concepts/categories)
  • Systems engineering (categories)
  • Systems modeling (facts/categories)
Modeling Interoperability

Languages Interoperability

The sudden, widespread, and impressive performance of generative languages like ChatGPT calls for a clarification of the relationships between languages and knowledge, beginning by a functional taxonomy of languages:

  • – Nominals: lexicon of words or groups of words attached to facts
  • – Modeling & Programming languages: syntax and semantics meant to be executed by computers
  • – Natural language: syntax, semantics, and pragmatics as used in humans communication

Then a taxonomy of of their use, in other words of understandings:

  • – Science (empiric) : Modeling & Programming languages meant to be aligned with observed facts (e.g. Simula)
  • – Formal (logic): Programming languages meant to preserve truth (e.g. Prolog)
  • – Generative (pragmatic): Inverse semantic parser meant to translate nominal inputs into natural language (e.g. ChatGPT)

Finally a taxonomy of languages as symbolic tools built on purpose:

  • Object oriented analysis and design, when the purpose is to manage symbolic representations (aka surrogates)
  • Goal oriented, when the purpose is to align enterprise organization and processes with business models
  • Fact oriented, when the purpose is to give structure and meaning to data
Integration & Interoperability of Modeling Paradigms

These taxonomies can be neatly aligned with the facets of the proposed ontological prism, ensuring their interoperability.

Modalities & Conceptual Interoperability

Semantic interoperability deals with the meaning of ontologies contents (what is known), modeling interoperability deals with their representation, conceptual interoperability deals with their status (how it is known), e.g.: the term used (nominal concept), a given pipe (fact), the pipe in Magritte painting (virtual pipe), a generic type (abstract pipe), corresponding types for actual pipes (actual concept) and referenced ones (category).

conceptual Interoperability

Ontological prisms can ensure conceptual interoperability by introducing built-in epistemic modalities :

  • Extensional modality: whether facts are observed, asserted, assessed, deduced, managed
  • Intentional modality: whether concepts are meant to be abstract (meanings, no instances), concrete (symbolic or physical instances), nominal (terms or labels, no meanings), or virtual (meanings set in the mind of beholders)
  • Ontological modality: how categories are apprehended
What is known (Ontologies) vs How it is known (Epistemology)

Epistemic modalities can be used to set knowledge engineering routines between: abstract concepts and categories (a), applied concepts and facts (b), and managed facts and categories (c).



%d bloggers like this: