Topic Maps
Topic Maps is an abstract model and an associated SGML or XML - based data format for the formulation of knowledge structures. Topic Maps 1999 were standardized as ISO standard ISO / IEC 13250 and later as XML Topic Maps ( XTM ) in XML formulated.
A topic map is the collection of knowledge about subjects, ie objects or subjects of description. Topics can be any things, such as people, places, events. A distinction is made between addressable subjects (which is all that can be stored in a computer ) and non- addressable subjects (anything that can not be stored in a computer, for example, you who read this page, the idea of freedom, etc.). Objects or subjects himself to and often can not be changed, if it is collected knowledge. Therefore Topics are used to represent the objects described in a topic map. Other components of a topic map are associations that describe the links between topics and occurrences that connect topics with documents eg on the WWW. One therefore speaks of the TAO (T for Topics, A for associations and for O occurrences ) of the Topic Maps. Furthermore, there are names and roles ( names of the function of a topic in an association ). A topic can contain multiple names, which are then assigned to a scope. In this way, a topic several names ( eg in different languages). Conversely, several topics can also have the same names in different scopes. Thus, a topic, which stands for the sun, for example, three names ( sun, sun, soleil ). Does the Scope English is just the name sun valid, etc. For example, a topic map appear in different languages by simply setting the correct scopes. Another example: The name duck can stand both for the pet as well as a faulty newspaper report. It can therefore there are two topics which have this name. In the first case the Scope wildlife could be in the second and journalism.
Topic Maps are based on the proven in human knowledge processing structures of index and thesaurus and develop them for the computerized needs further. This sets them apart from other approaches emanating from the background understandable for the computer formalization, see, eg, Resource Description Framework (RDF ) and Web Ontology Language (OWL). Because of their origin to avoid Topic Maps often not operated demands on the knowledge specification, which have been defined with a view of inference engines.
Topic Maps is intended to enable better navigation and search in Internet resources and other documents and used for the exchange of metadata. They have their roots in glossaries, classification systems ( for example, the topics of Procedure of the Open Directory ) and thesauri, but go in its expressiveness beyond it. So can be formulated ontologies, which can be mapped among others for the semantic web and RDF to Topic Maps. Proposals to implement topic maps with RDF and OWL, are present ( Cregan 2005).
In Tolog Query Engine is, for example, a system which makes it possible to derive new knowledge of a Prolog -like query.
In practice, only simple ( faceted ) classifications are modeled using topic maps often, so that it is independently a simplified subset was created with the exchangeable Faceted Metadata Language ( XFML ).
As a unified API for working with Topic Maps exists TMAPI.