An Analytic Aggregation-Based Ontology Alignment Approach with Multiple Matchers

• Published in: Advanced Techniques for Knowledge Engineering and Innovative Applications pp. 143 - 159
• Main Authors: Song, Fuqi, Zacharewicz, Gregory, Chen, David
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg 2013
• Series: Communications in Computer and Information Science
• Subjects: Alignment Algorithm; Analytic Aggregation; Multiple Matchers; Ontology Alignment
• ISBN: 9783642420160; 3642420168
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-642-42017-7_11

A critical aspect of providing data interoperability relies on ontological alignment and the successful semantic integration. Ontology alignment is being applied to these domains as a fundamental component. Many basic matching techniques have been proposed, however in order to adapt to the diverse sources of ontology and enhance the matching ability, the crucial point facing is how to choose and combine different alignment algorithms. In this chapter, an approach with multiple alignment algorithms is described. The algorithms are proposed from lexical, semantic and structural levels of source ontology. The algorithms are chosen by a comprehensive pre-defined strategy, one or more specific algorithms will be chosen according to the features of entities to be matched. To aggregate the different matching results dynamically and automatically, Analytic Hierarchy Process (AHP) is applied innovatively based on three similarity indicators, which reflect the essential features of source ontology. The results of the benchmarking experiment suggested that the approach has strong matching ability. It obtained high precision and promising evaluation results.