Querying MOF Repositories: The Design and Implementation of the Query Metamodel Language (QML)

Abstract

In a digital business ecosystem (DBE) information on the businesses and the services they provide may be described in terms of models and data which are used to se-mantically discover partners and services. The object management group (OMG) defines a four layered modelling architecture, the model driven architecture (MDA), which provides mechanisms for rapid development of modelling languages addressing domain problems using the meta object facility (MOF). MOF incorporates object oriented concepts and is a subset of UML. Furthermore, as users typically don't know how to make requests, the system has to be tolerant. The Query Metamodel Language (QML) is a language that exploits the Object Constraint Language (OCL) (which is very closely associated with UML and therefore MOF) to provide powerful query support on model repositories. This paper presents the motivation for QML along with its abstract syntax. It also introduces the framework for QML processing that incorporates information retrieval functionality and is used to formulate fuzzy queries using the extended boolean model. It describes how QML is integrated in the MOF architecture and how semantic expansion of queries and evaluation can be done in an effective way.