Microsoft Word 16 ok6408. doc

 
Scenario 2:
This scenario covers the case of a user that needs to discover geo-object 
sources that have classes that map to a specific geo-object class G of the 
reference ontology. In this scenario, the catalog is invoked through the 
DescribeObjectType operation, having G as one of its parameters. The 
expected answer is a XML representation of a set of triples of the form (G, p, 
R), where p is a property that maps G into range R, defined in the reference 
2
OWL - Web Ontology Language: 
http://www.w3.org/TR/owl-ref/

---

242 
ontology. In particular, p may represent a relationship between G and some 
class H defined in the ontology of one of the geo-object sources.
For instance, consider a user that needs information about “cities” in the 
world. The catalog will execute the query against the ontology repository to 
discover where to find objects of the class “city”. The answer of the 
DescribeObjectType operation will be metadata about “city”, as explained 
above, that enables the user to locate and access all geo-object sources, 
registered in the catalog, that contain information about cities. 
 
Scenario 3:
The catalog can be used to provide standard metadata to a user who is 
designing a new geo-object source that will participate in the federation. The 
usage scenario is very similar to the second scenario and will not be repeated 
here. 
5. ADDITIONAL 
CAPABILITIES 
All scenarios described in section 3.3 address the use of an OGOC to 
enable interoperability between geographic databases. They also suggest 
additional capabilities that can be added to the catalog enabling a 
cooperative behavior.
Indeed, the catalog may help users achieve their goals through request 
and answer transformations. This cooperative behavior will be useful to 
correct or reformulate user queries and system answers.
The cooperative behavior may be accomplished with the help of the 
ontologies stored in the catalog, enhanced with an inference engine.
Among other functions, a cooperative environment may (Hemerly et al., 
1993): 
• Correct a query; 
• Resolve ambiguities; 
• Generate alternatives to accomplish, as much as possible, the same 
purpose as the original query, when it fails; 
• Provide explanations about an answer; 
• Complement a query by supplying additional information. 
Consider a catalog about geographic places in the World. An example of 
an ambiguous query would be “Find Rio de Janeiro”. Figure 3 illustrates a 
city and a state with this name in Brazil. Figure 4 shows a second city, 
called “Rio de Janeiro”, this time in Peru. 

---