Onto-Med >> Theories >> GFO Part I Basic Principles

Overview

General Formal Ontology (GFO)

... world.¹

Philosophy is a source for inspiration, but its contribution to the solution of conceptual modelling problems seems to be limited.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... useful.²

A UML-profile - based on the GFO-module for functions - has been developed in the context of (13).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... development.³

For more details of the applications, cf. the Onto-Med website:
http://www.onto-med.de

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...Onto-Builder⁴

http://www.onto-builder.de

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... Web⁵

Cf. the World Wide Web Consortium (W3C) website, http://www.w3.org/2001/sw/

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... ontologies⁶

Cf. Open Biomedical Ontologies (OBO) at http://obo.sourceforge.net/ and the Gene Ontology (GO), http://www.geneontology.org/

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

....⁷

We stipulate these equivalences for practical reasons, since more subtle distinctions seem to be irrelevant in modelling practice. A deeper investigation of the relations between satisfiability, instantiation, and predication is a project for future research.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... language.⁸

We do not assume that every well-formed expression of a language expresses a category. Hence, the categorial expressions of a language form - in general - a proper subset of all its expressions. The investigation and understanding of categorial expressions is related to logic, linguistics and cognitive science which play a dominant role in conceptual modelling and computer-science ontologies.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... sets.⁹

The term hereditary finite set, for example, is an expression which denotes a category that can be predicated of sets. This category is an entity that is different from the set of all hereditarily finite sets.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... predicates.¹⁰

A full axiomatization is discussed further in Part II (Axiomatics and Ontology Language) of the report.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... generis.¹¹

The GFO approach to time is related to what P. Hayes calls the glass continuum (30). Furthermore, we advance and refine the theory of (1)

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... topoids.¹²

Again, we use ideas of Brentano (11) and Chisholm (18) for our theory.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... animals.¹³

The ability of recognizing a human face, for example, seems to be based on the existence of a persistant which is represented in our memory as a system of features. This persistant enables us to identify a face at a time-point by verifying this face as an instance of the persistant.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...13¹⁴

We emphasize that the construction of universals by cognition does not contradict philosophical realism. The idea that ``objective'' universals can be immediately mirrored without any intermediate step of conceptualization, i.e., without introducing concepts, would certainly be a kind of non-serious vulgar-realism.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... identity.¹⁵

A full elaboration of our approach to personal identity is much more complicated. It must consider the underlying process, the place of consciousness and will, and the dynamic interrelations between the persistant, the presentials, and the process.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... aspects.¹⁶

Thomas Aquinas tried to exclude processes that connect the presentials. For him, instantaneous entities - in our terminology presentials - are creations of God, and the continual connection between them is simply an illusion.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... boundaries¹⁷

Recall that ``coincident process boundaries'' refers to the fact that the respective time-boundaries coincide. It does not mean that the presentials themselves should coincide.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... complexity¹⁸

The categories of situations and situoids as discussed in sect. 11 are a first attempt to account for this in a systematic manner.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... presentials).¹⁹

This resembles the idea of ``indirect qualities'' in (40).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...states²⁰

Note here that this is a different notion of state. A substantiation of this is given below.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... Individual.²¹

In earlier texts these were referred to as ``properties'' and ``qualities''.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...gaerdenfors-p-2000-a²²

Note that the term ``property value'' here resembles Gärdenfors' notion of ``property'', our ``property'' his ``quality dimension''

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...masolo-c-2003-a²³

A quality space consists of all ``quales'' (our property values) of some ``quality'' (our property)

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... character.²⁴

Identifying the subcategories of the category to which relations belong, i.e. whether relations can exist as universals, concepts and the like, remains to be analyzed.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... roles²⁵

For convenience, ``role'' is used as an abbreviation for relational role this section.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...context.²⁶

Note that ``context'' here is just an auxiliary notion for introducing roles, instead of being presented in a profound ontological analysis.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...,²⁷

The literature provides fills and hasRole as other common terms for the plays relation.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... universals²⁸

Other terms in the literature are natural type (25), natural kind (63), phenomenon (54, p. 80), base classifier in UML (48, p. 194 ff.), and basic concept in (56).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...#tex2html_wrap_inline2923#²⁹

Depending on the reading of ``between'', one may identify with and with , respectively. For generality, we allow for a distinction of these time-boundaries.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...#tex2html_wrap_inline2961#³⁰

In general, there can be a special case that the end-point of and the starting point of coincide. In this case, the demolition would have no temporal extension, and is considered as a change instead of a process. It could be that - from the point of view of a certain granularity - this assumption is realistic.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... websites³¹

Each of these pages were available on 01.05.2006:
http://imaginis.com/breasthealth/staging.asp#what
http://www.cancer.org/docroot/ETO/content/ETO_1_2X_Staging.asp

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...#tex2html_wrap_inline3163#³²

Note that two different notions (here: tumor, tumor are commonly named tumor; this will be clear from below.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... Ontologies³³

http://obo.sf.net

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... relation³⁴

The annotation relation is primarily a database relation in biomedical ontologies. In particular, it relates genes or gene products to the categories of a biomedical ontology, meaning that a gene is somehow related to a category.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... development³⁵

For progress, see http://onto.eva.mpg.de.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... DOLCE³⁶

``Descriptive Ontology for Linguistic and Cognitive Engineering''

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... SUMO³⁷

``Suggested Upper Merged Ontology'', see homepage at
http://www.ontologyportal.org/

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... CYC³⁸

see http://www.cyc.com/cyc (commercial version) and http://research.cyc.com/ (research version)

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... 4D-ontology³⁹

http://www.tc184-sc4.org/wg3ndocs/wg3n1328/lifecycle_integration_schema.html

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... groups.⁴⁰

http://ontolog.cim3.net/cgi-bin/wiki.pl?UpperOntologySummit

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... relators⁴¹

Of course, in the formalization of DOLCE, relations are used and defined.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... p. 72)⁴²

This figure differs from the one in the book in having ,,Structure`` and ,,Situation`` as children of Nexus, instead of ,,Situation`` and ,,Execution`` (in this order). However, this figure corresponds to the text in the book. It is available from
http://www.jfsowa.com/ontology/toplevel.htm.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... categories.⁴³

We ignore the absurd category here, which is a subcategory of every category for Sowa. In contrast, GFO does not have a single intensional equivalent of the (extensional) empty set.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... ontology.⁴⁴

That means, there are discussions on time and space as well as on properties in (54), but it is difficult to determine whether these belong to the ontology actually promoted. With restriction to the lattice presented in figure 4, the statement is correct.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.