Ontologies have emerged as a result of investigations to overcome the problems presented in knowledge-based systems, mainly in knowledge acquisition. The above knowledge-based systems are mainly represented using production rules, semantic networks, etc. How ontologies are better than these ways of representing knowledge?
ontologies (semantic web ontologies) are published on the web in order to be reused... so they are not better than others knowledge representation model (you can do the same thing) they are only shareable, and you can describe an ontology according to others ones.
they are several kind of "ontology" in the semantic web: RDFS, OWL DL, OWL Full etc... but their main advantage is to be published and reused .
Along other advantages are machine accessibility . Machine can also do some job for you. what is the average age of person on the subcontinent who smoke?
. But if you have WWW ontology then you can assert these questions, semantic crawlers crawl the web process this knowledge represented in the form of ontology and provide results in compact manner.
One of the definitions of ontology is "a formal and explicit specification of a shared conceptualization". Let us adopt this definition.
I would like to emphasize the last part of the definition: "shared conceptualization". This is important for ontologies. Other knowledge representations ("non-ontologies") don't assume this constraint. Thus, in principle, other knowledge representations can represent any ad-hoc arbitrary conceptualization.
I think that all researchers agree that we cannot ensure that a given ontology captures a "shared conceptualization". However, when someone decide to develop an ontology, he/she tries to do the best approximation of the shared conceptualization.
Some people would like to say that a "knowledge representation" can be called an ontology if it tries to capture, in a formal and explicit way, the "invariant knowledge" of a domain, and this knowledge is an approximation of a shared conceptualization of this domain.
Some of us believe that reuse of the knowledge captured in an ontology, and the semantic interoperability, are consequences of the fact that the ontology captures a "shared conceptualization". If we represent arbitrary conceptualizations of the world, without the adoption of good tenets of knowledge acquisition, knowledge modeling, ontology engineering, ontological analysis, the resultant artifact hardly will be reusable.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs