Mediation

From Wsmx-WIKI
Jump to: navigation, search

Mission

Mediation in SEE aims at providing flexible mediation service at both data and process level.

The main focus on Data Mediation provides automatic transformation of data used in conversation between various parties based on ontology mappings. Additionally, techniques for ontology mappings optimization and global schema generation are investigated, together with their potential in query rewriting. As a support for all these mediation scenarios, a uniform mapping language is developed, with strong links and support towards automatic mappings generation.

The Process Mediation has the task of solving the communication (behavioral) mismatches that may occur during the communication between a requestor and a provider of a service. As in WSMO, the requestor is a WSMO Goal, while the provider is a Semantic Web Service, the Process Mediator’s task is be to accommodate the mismatches between the goal’s requested Choreography and the SWS’s choreography.


Team

Leader:

   Adrian Mocan

Members:

   Emilia Cimpian
   Richard Pöttler
   François Scharffe
   Omair Shafiq

Observer:

   Mick Kerrigan

[ Meetings ] [ News ] [ Events ] [ Presentations ]


Conceptual Overview

Data Mediation - Instance Transformation

The Data Mediation service in SEE aims to provide a solution to solve the heterogeneity problems that can appear at the data level. As all messages in WSMX are semantically described in WSML, the data to be mediated is described in terms of ontologies, i.e. data consists of ontology instances.

In this context, the heterogeneity problems at the data level appear when the requester and the provider of a service use different ontologies to conceptualize their domain. As a consequence, data has to be transformed from terms of one ontology (e.g. requester’s ontology) into terms of the other ontology (e.g. provider’s ontology). Due to the fact that these transformations are taking place during run-time the whole process has to be completely automatic. The data mediator component in WSMX achieves this by relying on a set of mappings (semantic relationships) between the source and target ontology identified during design-time and stored in a persistent storage.

The mappings are in fact logical rules that are executed during run-time by a reasoner component against the incoming data, to output data as required by the target party. There are several ways (languages) of representing these rules, depending of the reasoning support available. In order to encourage interoperability between various mediation systems and to allow a flexible and an easy management of these mappings, a language independent format (called the Abstract Mapping Language) is used. As a consequence, each time a set of such mappings have to be used in a concrete scenario (as the instance transformation in WSMX) the mappings have to be “grounded” to a concrete ontology representation language (in our case WSML). The grounding not only transforms the mappings in an executable form, but also associate them a formal semantics, a meaning in respect with the concrete representation language and the mediation scenario to be used in.

The WSMX Data Mediation service is a data mediation engine capable of performing instance transformation for given pairs of ontologies based on a given set of mappings. These mappings are represented as statements in an Abstract Mapping Language which assures an ontology representation language neutrality. By applying different grounding mechanism, the same set of abstract mappings can be used in different mediation scenarios using different reasoning systems.


Data Mediation - Query rewriting

While instance transformation is generally effective there are some cases where we directly want to query the mediated data source. In that respect we need to transform the query given the mapping between the two sources. The WSMX data mediation module will therefore include a query rewriting engine answering these needs.


Process Mediation

Process level mediation deals with solving the interaction mismatches. Every semantic Web service has a specific choreography that describes they way in which the user should interact with it. Similarly, every requester of a service describes the way he wants to interact with a Web service by defining its own choreography. These choreographies describe semantically the control and data flow of messages that every partner can exchange. In cases where the choreography of the requestor (goal choreography) and the choreography of the Web Service do not match, process mediation is required. The Process Mediation component in WSMX operates based on the two choreographies, and it is responsible for resolving mismatches between the choreographies (often referred to as public processes). These mismatches can appear not only when the requestor and the provider of a service use different conceptualizations of a domain (in which case data mediation is required), but also if they have different requirements for the message exchange pattern to be followed. Essentially this means that one of them expects to receive/send messages in a particular order while the other one has a different messages or a different message order that doesn’t match. The role of the process mediator is to retain, postpone, rebuild or create messages that would allow the communication process to continue [Cimpian and Mocan, 2005].


Mapping Language

When mediating between heterogeneous data sources describe using possibly heterogeneous formalisms like XML, OWL or relational schemas, a common language to describe the semantic relationships between the schemas becomes necessary. The WSMX data mediation editor is based on top an abstract mapping language [Scharffe & de Bruijn, 2005] which is used as an exchange format between the different mediation component.


Mediation support for Triple Space Computing

While focusing on scalability and openness of Triple Space Computing middleware, the possibility of heterogeneity among different users communicating over Triple Space is very likely to arise. The aim is to provide Triple Space Computing with mediation support for easy integration of data, information, knowledge, processes, applications, and businesses which is one of the core issues in interoperability due to the heterogeneity of the interacting entities.

Mediation Engine in Triple Space Kernel is concerned with handling heterogeneity by resolving possibly occurring mismatches among different triples. There can be a possibility that different TSC participants communicating over Triple Space, containing different data models which should be resolved by the process of mediation. So, a RDF instance in a RDF schema of one TSC participant is needed to be represented in the RDF schema of the other TSC participant without altering or loosing the semantics. For this reason, a mapping language is needed that specifies how to transform RDF triples according to different RDF Schemas of different participants. The mediation rules are to be specified at design time which will be processed by a mediation engine in the TS Kernel at runtime in order to carry out the mediation during the communication among TSC participants.

Abstract Mapping Language (independent of any ontology languages) has been used to specify the mapping rules. The mediation management APIs for users that would allow them to add, remove or modify mapping rules at design time. Grounding of the mapping rules to RDF has also been provided so that the mapping rules could be stored over Triple Space. The architecture of the mediation engine has been designed along with its bindings with the Triple Space Kernel to execute mediation rules at run time, and bindings with Coordination Layer to store and retrieve the mapping rules in Triple Space.

Mediation in TSC - Time Plan:

- Investigating the use of Abstract Mapping Language - Done

- Defining the Mediation Engine APIs - Done

- Mediation Engine Architecture - Done

- Mediation Engine bindings with Operations and Coordination layer - Done

- RDF schema for grounding mapping rules in Abstract Mapping Language - In progress (Deadline 20 September 2006)

- Implementation - To be started (Deadline = March 2007)

Implementation

Data Mediation - Instance Transformation

Available for download both as a stand-alone component and part of the WSMX architecture at http://sourceforge.net/projects/wsmx/

Process Mediation

Available for download together with the WSMX architecture at http://sourceforge.net/projects/wsmx/

Mapping Language API

Available for download from http://sourceforge.net/projects/mediation/

Future Work (Road Map)

Publications

[Shafiq et al., 2006] Omair Shafiq, Francois Scharffe, Reto Krummenacher, Ying Ding, Dieter Fensel, Data Mediation Support for Triple Space Computing. 2nd IEEE International Conference on Collaborative Computing (CollaborateCom 2006), 17-20 November 2006; Atlanta, GA, USA.

[Mocan et al., 2006] Adrian Mocan, Emilia Cimpian, Mick Kerrigan: Formal Model for Ontology Mapping Creation, The 5th International Semantic Web Conference (ISWC 2006), November 2006, Athens, USA

[Cimpian et al., 2006] Emilia Cimpian, Adrian Mocan, Michael Stollberg: Mediation Enabled Semantic Web Services Usage, 1st Asian Semantic Web Conference (ASWC2006), September 2006, Beijing, China

[Stollberg et al., 2006] Michael Stollberg, Emilia Cimpian, Adrian Mocan, Dieter Fensel: A Semantic Web Mediation Architecture, Canadian Semantic Web Working Symposium (CSWWS 2006), June 2006, Québec city, Canada

[Mocan & Cimpian, 2005] Adrian Mocan, Emilia Cimpian: Mapping Creation Using a View Based Approach, 1st International Workshop on Mediation in Semantic Web Services (Mediate 2005), December 2005, Amsterdam, Netherlands

[Stolberg et al., 2005] M. Stollberg, E. Cimpian, D. Fensel: Mediating Capabilities with Delta-Relations, 1st International Workshop on Mediation in Semantic Web Services (Mediate 2005), December 2005, Amsterdam, the Netherlands.

[Cimpian & Mocan, 2005] Emilia Cimpian, Adrian Mocan: WSMX Process Mediation Based on Choreographies, 1st International Workshop on Web Service Choreography and Orchestration for Business Process Management (BPM 2005), September 2005, Nancy, France.

[Scharffe & de Bruijn, 2005] Francois Scharffe, Jos de Bruijn: A Language to specify Mappings between Ontologies, IEEE Conference on Internet-Based Systems (SITIS6), December 2005, Yaounde, Cameroon.