AG Netzbasierte Informationssysteme (Networked Information Systems) | Freie Universitšt Berlin

MOMA Framework

FU Berlin

MOMA Objectives.

Due to its open design, a fully developed Semantic Web will contain numerous, distributed and ubiquitously available ontologies. Users will be able to choose among them to allow a mediated access to Web information, or to integrate or transform them to application-specific, customized models. Further on, the next generation of Web Services will apply ontologies to describe service capabilities and to mediate inter-process communication. A fundamental requirement for the realization of this vision are proved and tested ontology matching algorithms, which are able to deal with the heterogeneity of current ontological sources available on the Web. The importance of ontology matching is emphasized by its implication in most of the phases of an ontology management process, be that ontology merging, mapping or evaluation. Though containing valuable ideas and techniques existing matching algorithms cannot to be used optimally in ontology matching tasks as those envisioned by the Semantic Web community, mainly because of the inherent dependency between particular algorithms and ontology properties such as size, representation language or underlying graph structure, and because of performance and scalability limitations as well as lack exhaustive testing in real world scenarios. In order to cope with the first problem we have designed a Metadata-based Ontology MAtching Framework (MOMA Framework) which exploits the valuable ideas embedded in current matching approaches but in the same time accounts for their limitations - for specific input ontologies it optimizes the matching results by eliminating unsuitable candidate matching methods. The MOMA Framework will allow both manual (MOMA-M: based on Analytic Hierarchy Process) and automated (MOMA-A: rule-based) detection of suitable matching approaches for the processing of a pre-defined ontological input.

Survey of Ontology Matching Systems

In order to provide the MOMA Framework to support for its decision about the current relevant and suitable matching algorithms we need information about a particular matcher. For this purpose we have developed a questionnaire that helps us get data regarding various matching approaches. This collected information will serve as a knowledge basis for our selection tool.

We would like to ask you if you could contribute to our survey the matching data you have regarding approach developed by your group. In turn your approach will be present within a list of potentially suitable matching approaches. When the tool for approach selection is in use, your approach might be chosen for a particular situation and used in various domains, applications and tasks. On the other hand, your contribution will of course be a great help for us as a first step in the development of the recommendation framework which aims to support young and inexperienced Semantic Web users and application developers.
The online survey can be found under: http://matching.ag-nbi.de

©2006 Freie Universitšt Berlin