Syntactic discovery of web service
Nowadays, thanks to the rapid development of semantic web, many web services carrying semantic meaning; i.e. semantic web service; are available. There have been a number of researches conducted to discover semantic web, i.e. between a semantic requester and semantic provider. However, there are st...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2009
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/16850 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | Nowadays, thanks to the rapid development of semantic web, many web services carrying semantic meaning; i.e. semantic web service; are available. There have been a number of researches conducted to discover semantic web, i.e. between a semantic requester and semantic provider. However, there are still very limited studies on how to match a semantic web service against a syntactic web service – the previous generation of web service. This is because the difference in nature of the two services makes the discovery process complicated. If the matter is solved, discovery system can fully exploit all the web services available online, no matter what type of service they are. This project proposes a solution to this matter.
Looking in more details, this project will propose mechanism for matching WSDL against WSDL, and using this result as an additional aid to the main purpose: matching WSDL against OWLS (or OWL-S against WSDL). In WSDL vs. WSDL matching, the system will look into their operations, text descriptions, inputs and outputs contained in the services to evaluate their similarities. Since they are syntactic information, their approximation can be estimated using string matching algorithm available from many researches. The same mechanism applies to matching an OWL-S service and WSDL service. The operation, input, output of the OWL-S service profile are matched against those of WSDL service (1).
However, this approach has not utilized the semantic aspect. Semantic web service describes information by concept. This concept, defined in ontology, has relationships such as super, sub and equivalent concepts. Each of these concepts describes a piece of meaningful information. Together, they can demonstrate alternative meaning in the real world. Understanding this nature, during the matching process, information from WSDL service must be matched with not only concept in OWL-S service but also with its other relationships, in order to explore the completely meaning of any concept (2).
The discovery system for WSDL against OWL-S will combine both ideas from (1) and (2) to produce an effective mechanism for matching a syntactic service against a semantic service. To ease the implementation, assumptions are made and parameters are initialized with certain values. Along the development process, these parameters are adjusted according to the testing results. Up to this point, the project has proved to give a positive outcome in evaluating syntactic and semantic similarity. |
---|