Ontology Matching with CIDER: evaluation report for OAEI 2011

Abstract

CIDER is a schema-based ontology alignment system. Its algorithm compares each pair of ontology terms by, firstly, extracting their ontological contexts up to a certain depth (enriched by using lightweight inference) and, secondly, combining different elementary ontology matching techniques. In its current version, CIDER uses artificial neural networks in order to combine such elementary matchers. In this paper we briefly describe CIDER and comment on its results at the Ontology Alignment Evaluation Initiative 2011 campaign (OAEI’11). In this new approach, the burden of manual selection of weights has been definitely eliminated, while preserving the performance with respect to CIDER’s previous participation in the benchmark track (at OAEI’08).