Solution DERI

From SWS Challenge Wiki

The DERI solutions were presented in SWS-Challenge phase two, three, and four, which covers various service-oriented issues/solutions in a B2B integration scenario:-

  • B2B Integration on Semantic Web Services (SWS), the work presented, introduces global integration architecture based on WSMX semantic web service execution environment, WSML Web service modelling language, and underlying conceptual model WSMO.
  • Web service discovery that extends the design and implementation of WSMX. The proposed discovery mechanism introduces “contracting interface”, interface that can be queried to provide run-time instance data generally unavailable in the description of Web services. The discovery mechanism presented, increases the automation level of goal-capability match.
  • The discovery work presented in phase two, was applied to constraint based Web service composition. Inconsistency and incompleteness are the inherited nature of any true distributed systems like World Wide Web (WWW). Work presented in phase 3, shows the importance of run-time data fetching during composition of Web services to enrich the incomplete Web service description.

Contents

[edit]

B2B Integration on Semantic Web Services (SWS)

The integration scenario includes standard (e.g. RosettaNet) and propriety based systems where SWS technologies were applied to data and process interoperability. The work demonstrated can be categorized into:

  • Data grounding and Mediation
    • XML schema (RosettaNet) <= => WSML
    • XML schema (Propriety systems) <= => WSML
    • WSML(RosettaNet) <= => WSML(Propriety systems) [Mapping]
  • Web service description grounding and process mediation
    • WSDL(RosettaNet) <= => WSML service description
    • WSDL(Propriety systems) <= => WSML service description
    • WSMX, a semantically-enabled integration middleware facilitates the overall integration of heterogeneous systems, including discover, composition, and choreography of Web services.
  • The work presented reflects overall enterprise integration scenario that shows how data and process mediation can be applied between heterogeneous services.

Paper: T. Hasselwanter, P. Kotinurmi, M. Moran, T. Vitvar, M. Zaremba: WSMX: a Semantic Service Oriented Middleware for B2B Integration, In proceedings of the 4th International Conference on Service Oriented Computing , Springer-Verlag LNCS series, December, 2006, Chicago, USA.

Presentation: WSMX: a Semantic Service Oriented Middleware for B2B Integration

[edit]

Web Service discovery

  • In phase three, a Goal based approach was demonstrated for the task of discovering Web services. Goal (requirements) of services requester and Capability (what it can offer) of the provider are semantically described to achieve (total or partial) automation in complete Web service discovery process.
  • A discovery approach is introduced to deal with incomplete service description, called "contracting interface", this Web service interface is queried at run-time to fetch additional data that is not a part of Web service description.

Paper: T. Vitvar, M. Zaremba, M. Moran: Dynamic Service Discovery through Meta-Interatctions with Service Providers, In Proceedings of the 4th European Semantic Web Conference (ESWC2007) , Springer-Verlag LNCS series, June, 2007, Innsbruck, Austria.

M. Zaremba, T. Vitvar, M. Moran: Towards Optimized Data Fetching for Service Discovery, The 5th IEEE European Conference on Web Services (ECOWS) , IEEE CS Press, November, 2007, Halle, Germany.

Presentation: Dynamic Service Discovery through Meta-Interatctions with Service Providers

[edit]

Web Service composition

  • In Phase four, various semantic (information, functional, and behaviour) definitions provided at WSMO-Lite included as conceptual base of the work presented.
  • Two extensions to discovery work presented at phase three are introduced,
    • Goal decomposition (sub-goal and capability matching)
    • Managing knowledge base for all sub-goal/goal-capability matching information.
  • Solution was based on the asumption that Web services composition are constraint based where invocation order is unimportant. The focus of the work was to show, when constraints are run-time data dependent, how to fetch these instance data and use them during service composition.

Paper: M. Moran, T. Vitvar, M. Zaremba: SWS-Composition Towards Constraint-Based Composition With Incomplete Service Descriptions , The International Workshop on Service Composition and SWS Challenge in conjunction with the IEEE/WIC/ACM International Conferences on Web Intelligence (WI 2007) , IEEE CS Press, November, 2007, Silicon Valley, USA.

[edit]

Contact

Team Members:

Retrieved from "http://sws-challenge.org/wiki/index.php/Solution_DERI"