Matching large schemas: Approaches and evaluation

Abstract

Current schema matching approaches still have to improve for large and complex Schemas. The large search space increases the likelihood for false matches as well as execution times. Further difficulties for Schema matching are posed by the high expressive power and versatility of modern schema languages, in particular user-defined types and classes, component reuse capabilities, and support for distributed schemas and namespaces. To better assist the user in matching complex schemas, we have developed a new generic schema matching tool, COMA++, providing a library of individual matchers and a flexible infrastructure to combine the matchers and refine their results. Different match strategies can be applied including a new scalable approach to identify context-dependent correspondences between schemas with shared elements and a fragment-based match approach which decomposes a large match task into smaller tasks. We conducted a comprehensive evaluation of the match strategies using large e-Business standard schemas. Besides providing helpful insights for future match implementations, the evaluation demonstrated the practicability of our system for matching large schemas. r

Ten Challenges for Ontology Matching

“…Specifically, for certain and quite frequent in practise cases, e.g., when matching formula is Horn, satisfiability became resolved in linear time, while standard SAT solver would require quadratic time, see [40] for details. Beside S-Match, several other groups, for example, Falcon [44] and COMA++ [13], have started addressing seriously the issues of performance. However, this fact cannot be still considered as a trend in the field, see, e.g., the results of the anatomy track of OAEI-2007 [22], where only several systems, such as Falcon, took several minutes to complete this matching task, while other systems took much more time (hours and even days).…”

“…This increases the complexity of the previous problem by allowing to put several matchers together and to combine them adequately. So far, only design time toolboxes allow to do this manually [13]. Another approach involves ontology meta-matching [50], i.e., a framework for combining a set of selected ontology matchers.…”

“…So far, there have only been few studies on how to involve users in ontology matching. Most of these efforts have been dedicated to design-time matcher interaction [13,66].…”

Measuring the Quality of an Integrated Schema

“…In this section, we present the evaluation results of the following schema matching tools: COMA++ [7,8] and Similarity Flooding (SF) [9,10]. These tools are described in the next section (see section 5).…”

On the Use of Description Logic for Semantic Interoperability of Enterprise Systems

“…Mapping is defined by [17] Automatic or semi-automatic mapping may use some mapping tools such as FCAMerge [20], IF-map [21], GLUE [22], COMA++ [23]. These automatic or semiautomatic mapping tools achieve accurate mapping results under the conditions of two ontologies defined in natural-language descriptions which are at the conceptual level.…”