Web servis koleksiyonlarının tam otomatik anlamlandırılması

The definition and use of a semantic layer in Web Services (WS) descriptions is a prerequisite to the automation of several important operations such as WS composition and mining. For this reason, in the past years, many approaches have been proposed to either represent such high level information, or take advantage of it. In order to be properly tested and compared, these tools must be applied to an appropriate benchmark, taking the form of a collection of semantic WS descriptions. However, all of the existing publicly available collections are limited in terms of size or realism, leading to unreliable results. Large real-world syntactic (WSDL) collections exist, so an appropriate benchmark could be obtained through their semantic annotation. Due to the number of operations to process, performing this task manually would be costly in terms of time and efforts, though. A better solution would therefore be to use an automatic tool. With this motivation, in this work we propose a new fully automatic semantic annotation tool, called MATAWS, designed to process WS descriptions. The resulting tool takes advantage of the latent semantics present not only in the parameter names, but also in the data type names and structures. Concept-to-word association is performed by using Sigma, a mapping of WordNet to the SUMO ontology. The design of MATAWS is data-driven, based on the properties of ASSAM Full Dataset, the largest collection of real-world syntactic WS descriptions we could find. After having described in details our annotation method, we apply it to this collection and assess its efficiency. We also apply and evaluate MATAWS on another WS collection prepared by ourselves, in order to prove our tool performs well on new, independent data. Keywords: Web Service, Semantic Web, Semantic Annotation, Ontology, WSDL, OWL-S, SUMO, WordNet.