|
WSRDGen is a Java class library that provides an API to generate a global
semantic description from a set of semantic web services descriptions written
in SAWSDL. This class library parses and extracts semantic concepts
annotating the different SAWSDL descriptions elements (Interface, Operation,
Input and Output). The resulting hashMap of each
element containing the different WSRD concepts as well as their occurrences are
the inputs for building the clouds of potential mean concepts. The clouds, in
turn, are stored in hashMaps representing all
ontology concepts.
Finally, we select the
semantically most similar concepts using two different techniques: strong
(also called hard) and weak (also called fuzzy). The first one chooses the
concept having the highest value. However, the second consists in choosing
more than one concepts according to a specific algorithm.
|
|
The classes within org.wsrd.generatingWSRDFile package, namely, HardWSRDFile and FuzzyWSRDFile represent the central classes for this API.
Both of them generate a WSRD model file but with two different methods. In
fact, the former uses the strong reduction method. However, the latter uses
the weak one.
|
|
WSRDGen binariesExternal
libraries
|
|
Here, we present a basic exemple using the strong reduction method in order to
generate a WSRD model file from a set of 10 Web services descriptions.
|
|
Mohamed[DOT]Sellami[AT]it-sudparis.eu
|