Schema and ontology matching with COMA++

Abstract

We demonstrate the schema and ontology matching tool COMA++. It extends our previous prototype COMA utilizing a composite approach to combine different match algorithms [3]. COMA++ implements significant improvements and offers a comprehensive infrastructure to solve large real-world match problems. It comes with a graphical interface enabling a variety of user interactions. Using a generic data representation, COMA++ uniformly supports schemas and ontologies, e.g. the powerful standard languages W3C XML Schema and OWL. COMA++ includes new approaches for ontology matching, in particular the utilization of shared taxonomies. Furthermore, different match strategies can be applied including various forms of reusing previously determined match results and a so-called fragmentbased match approach which decomposes a large match problem into smaller problems. Finally, COMA++ cannot only be used to solve match problems but also to comparatively evaluate the effectiveness of different match algorithms and strategies.

A Functional Model for Dataspace Management Systems

“…12 Bioinformatics use case: Maintenance by evolving the mediated schema in response to changes to the addition of OMIM and Ensembl to the dataspace 1. a view of dataspace management systems as second-generation data integration platforms in which pervasive automation is deployed to push down costs and user feedback is gathered and used to increase result quality; 2. a view of the life cycle of a dataspace that captures, albeit coarsely, the functionality of a growing body of literature (which we have most recently surveyed in [33]), is consistent with (1), and comprising initialization, use, maintenance and improvement stages; 3. a conceptualization of the functional architecture of dataspace management systems that is consistent with (1) and explains more precisely than has been done so far in what ways they are related to classical database, data integration and model management systems; 4. a formalization of the functional model for dataspace management systems as a many-sorted algebra that is consistent with (3), and, while building on a history of advances by data integration and model management researchers, gives crisp contours to the notion of a dataspace, contours that had been hitherto only ambiguously and vaguely drawn; 5. examples of algebraic programs that illustrate how the functional model in (4) can capture and support the various stages of the dataspace life cycle in (2) as well as a more extended use case in bioinformatics that shows how important practical scenarios can be supported.…”

“…Broadly construed to encompass systems that are primarily concerned with matching management or with mapping management, as well as the full gamut of model management capabilities, the area has produced impressive research systems such as COMA [19,4], Clio [55,56,16,34] (some of whose contributions have been incorporated into DB2), AutoMed [12,11,58], Rondo [54,53,51], GeRoMe [40] and MISM/MIDST [2,3], among others.…”

“…, a n resulting from a sequence of pairwise matching operations can be aggregated (e.g., by simple union); and (d) whether the associations are, as suggested above, between sets of constructs or between single constructs. In implementing this operation, one might use one or many algorithms, and in the latter case, one might use different strategies for aggregating the scores returned by individual algorithms (see, e.g., [19,4]). f := inferCorrespondences(a) Given a set a of associations, a set f of semantic correspondences can be generated from it.…”

Automatically Mapping and Integrating Multiple Data Entry Forms into a Database

“…The limitations of our approach include the missing correspondences and syntactic nature of the merging process. In future, we intend to exploit more sophisticated schema matching techniques [6] and apply an ontology to resolving semantic heterogeneity.…”

Online Social Network Profile Linkage

“…Surveys [8,9]. review the various approaches, including named attributes computations [5], schema mapping [2,17] and duplicate detection in hierarchical data [10], all which inform the construction of profile linkage techniques.…”