European Journal of Scientific Research - EuroJournals

380 Benharzallah Saber and Kazar Okba
evolution, a division and an easier inference of ontologies [27]. The language used to interrogation the
semantic mediation system is SOWL [28].
We adopt that ontology described the concepts of a domain while the schema describes the
structure of a database, and the establishment of bonds between them makes it possible to data
integrate while being based on their significance and not on their structure. Our approach uses the
OWL to represent all information concerning the semantic mediation of the information systems. OWL
is used like
• A common representation model of the schemas: our approach is based on work which was
completed to convert schema represented on the relational model, object or XMLSchema [21]
[22] [23] towards OWL model.
• A common representation model of ontologies: to clarify the concepts of the schema compared
to the application domain of the cooperation system.
• A representation language of the links schema-ontology: to link a schema to its ontology.
• A representation language of the mapping rules: to link a consuming schema system to
suppliers schemas systems, in the case of the schema mediation
3.6. Queries processing
The query processing breaks up into several steps, and during this process, the multi-agents system
uses a set protocol. The principal steps are:
1. Static query resolution
Step 1: query validation the IAC checks the validity of the query. i.e. is this which is written in
schema mediation terms or not.
Step2: query decomposition: the request is broken up at a recombining query of the results and sub
queries intended for the intelligent agent suppliers (IAS) who contain data necessary to the execution
of the query. The decomposition of the query is done by the use of the mapping rules. The IAC applies
the cooperation protocol of static query resolution.
Step3: recombining of the results: the intelligent agent consuming (IAC) execute the recombining
query of the results.
2. Dynamic query resolution
The dynamic resolution makes it possible to take into account the appearance of new IASs. The
principal steps are:
Step 1: query reformulation the IAC reformulates the query in its ontology terms by using the links
schema-ontology which is in its knowledge base.
Step 2: Transmission of the reformulated query the IAC applies the cooperation protocol of
dynamic query resolution. So it transmits the reformulated query to the routing agent which is nearest
semantically. And this last sends it to all IASs of its net contacts.
Step 3: Semantic evaluation of the reformulated query: Each IAS answers according to its capacity
to treat the query:
a. To compare elements of the query with its ontology. The elements of the query and its ontology
are compared using a semantic distance. The identified elements as equivalent are retained.
b. The query is rewritten in terms of the equivalent elements of its ontology (then interpreted on
its schema) to take of them into account the semantic conflicts of values (each intelligent agent
has library of functions for the conversion of the types).
c. The answer is sent then to the routing agent, by indicating the manner of treating the query, for
this last can build recombining queries of the results.
If no IAS answers, the routing agent sends the query to the other routings agents of other
domains and if there are answers the routing agent updates its net contacts.