Classifications

A classification is a type that describes a method, which divides a set of objects completely into distinct subsets by means of categories. Hierarchical classification might be defined by defining sub categories for each category etc. Well-formed classifications do have the same nesting level for each category.

When a classification refers to a type or is being defined in the context of an object type, the classification is called typed classification. Typed classifications can apply on objects of the type associated with the classification, only.

Classifications defined as features of a type are considered automatically as typed classifications.

Classifications are referenced as type for characteristics (properties) in order to define classifying characteristics. By assigning categories to individual object, objects are associated with this category.

In order to define additional characteristics for categories of the classification, the classification may contain to a number of properties (characteristics).

In [1087], there is a limited support for categories (type of characteristic). [ODMG] supports simple classifications as enumeration. [ERM] does not support classifications, but those might be implemented as tables and table relations. [UML] supports classifications as enumerations. [UDT] considers classifications as one way for providing a P3 database schema.

A classification supports a list of categories. The set of categories is supposed to divide a given object collection completely into distinct subsets.

A classification may define a number of properties (characteristics) for it's categories. Each property defined for the classification becomes a value in the subordinated categories.

Classifications may refer to an object type, which describes the kind of objects the classification may apply on.