• Гідрологія і Гідрометрія
• Господарське право
• Економіка будівництва
• Економіка природокористування
• Економічна теорія
• Земельне право
• Історія України
• Кримінально виконавче право
• Медична радіологія
• Методи аналізу
• Міжнародне приватне право
• Міжнародний маркетинг
• Основи екології
• Предмет Політологія
• Соціальне страхування
• Технічні засоби організації дорожнього руху
• Товарознавство продовольчих товарів

Motivations

We characterize the meaning of the term “ontology” to include a catalog of terms used in a

domain, the rules governing how those terms can be combined to make valid statements about

situations in that domain, and the “sanctioned inferences” that can be made when such statements

are used in that domain. In every domain, there are phenomena that the humans in that domain

discriminate as (conceptual or physical) objects, associations, and situations. Through various

language mechanisms, we associate definite descriptors (e.g., names, noun phrases, etc.) to that

phenomena. In the context of “ontology,” we use the term “relation” to refer to a definite

descriptor that refers to an association in the real world. We use the term “term” to refer to a

definite descriptor that refers to an object or situation-like thing in the real world. In an ontology,

we try to catalog the descriptors (like a data dictionary) and create a model of the domain, if

described with those descriptors. Thus, in building an ontology, you must produce three

products. You have to catalog the terms, capture the constraints that govern how those terms can

be used to make descriptive statements about the domain, and then build a model that when

provided with a specific descriptive statement, can generate the “appropriate” additional

descriptive statements. By appropriate descriptive statements we mean (i) because there are

generally a large number of possible statements that could be generated, the model generates only

that subset which is “useful” in the context, and (ii) the descriptive statements that are generated

represent facts or beliefs that would be held by an intelligent agent in the domain who had

received the same information. The model is then said to embody the “sanctioned inferences” in

the domain. It is also said to “characterize” the behavior of objects and associations in the

domain. Thus, an ontology is similar to the now familiar data-dictionary, plus a grammar, plus a

model of the behavior of the domain.

Another characterization of the meaning of “ontology” is given in the following excerpt from

Tom Gruber (see [Gruber 93] also):

The word “ontology” seems to generate a lot of controversy in discussions about

AI. It has a long history in philosophy, in which it refers to the subject of

existence. It is also often confused with epistemology, which is about knowledge

and knowing.

In the context of knowledge sharing, I use the term ontology to mean a

specification of a conceptualization. That is, an ontology is a description (like a

formal specification of a program) of the concepts and relationships that can exist

for an agent or a community of agents. This definition is consistent with the usage

of ontology as set-of-concept-definitions, but more general. And it is certainly a

different sense of the word than its use in philosophy.

Ontologies are often equated with taxonomic hierarchies of classes, class

definitions and the subsumption relation, but ontologies need not be limited to

these forms. Ontologies are also not limited to conservative definitions, that is,

definitions in the traditional logic sense that only introduce terminology and do

not add any knowledge about the world (Enderton, 1972). To specify a

conceptualization, one needs to state axioms that do constrain the possible

interpretations for the defined terms.

Pragmatically, a common ontology defines the vocabulary with which queries and

assertions are exchanged among agents. Ontological commitments are

agreements to use the shared vocabulary in a coherent and consistent manner. The

agents sharing a vocabulary need not share a knowledge base; each knows things

the other does not, and an agent that commits to an ontology is not required to

answer all queries that can be formulated in the shared vocabulary.

Any domain with a determinate subject matter has its own terminology, a distinctive vocabulary

that is used to talk about the characteristic objects and processes that comprise the domain. A

library, for example, involves its own vocabulary having to do with books, reference items,

bibliographies, journals, and so forth. Similarly, semiconductor manufacturing has its own

language of chips, wafers, etchants, designs, and so on. The nature of a given domain is thus

revealed in the language used to talk about it. Clearly, however, the nature of a domain is not

revealed in its corresponding vocabulary alone; in addition, one must (i) provide rigorous

definitions of the grammar governing the way terms in the vocabulary can be combined to form

statements and (ii) clarify the logical connections between such statements. Only when this

additional information is available is it possible to understand both the natures of the individuals

that exist in the domain and the critical relations they bear to one another. An ontology is a

structured representation of this information. More exactly, an ontology is a domain vocabulary

together with a set of precise definitions, or axioms, that constrain the meanings of the terms in

that vocabulary sufficiently to enable consistent interpretation of statements that use that

vocabulary.

Taken by itself, it may seem that there is not much difference between an ontology and a data

dictionary. However, a data dictionary is typically just a compendium of terms together with

definitions for the individual terms stated in natural language. By contrast, the grammar and

axioms of an ontology are stated in a precise formal language with a very precise syntax and a

clear formal semantics (see Section 4.2). Consequently, ontologies are, in general, far more

rigorous and precise in their content than a typical data dictionary (and, hence, more so than a

typical data “encyclopedia,” because an encyclopedia is just a collection of related data

dictionaries). Ontologies also tend to be more complete as well: relations between concepts and

objects in a domain, and constraints on and between domain objects, are made explicit rather

than left implicit, thus minimizing the chance of misunderstanding logical connections within the

domain. A data dictionary, by contrast, generally relies upon an intuitive understanding of the

terms in question and the logical connections between the concepts and objects they stand for.

This works well enough in small restricted domains. But when information systems span

organizational, geographic, and enterprise boundaries, problems arise. The traditional approach

is problematic for several reasons, not the least of which is that different persons in different

domains might understand the same term subtly different but important ways that are not

uncovered in a natural language definition (which can lead to inconsistent interpretations of the

same term across different contexts), and so forth. The regimentation of an ontology of the

involved domains in a canonical language helps to avoid this problem. Furthermore, the

discipline of expressing the ontology information in a formal language enhances the skills

necessary for extracting the information, in particular, the ability to abstract from particular

objects to the kinds of which they are instances, from particular connections to the relations such

instances stand in generally, and from particular behaviors to the constraints that bind instances

of various kinds together logically within the domain.

Переглядів: 457