Using Partial Reference Alignments to Align Ontologies

In different areas ontologies have been developed and many of these ontologies contain overlapping information. Often we would therefore want to be able to use multiple ontologies. To obtain good results, we need to find the relationships between terms in the different ontologies, i.e. we need to align them. Currently, there already exist a number of ontology alignment systems. In these systems an alignment is computed from scratch. However, recently, some situations have occurred where a partial reference alignment is available, i.e. some of the correct mappings between terms are given or have been obtained. In this paper we investigate whether and how a partial reference alignment can be used in ontology alignment. We use partial reference alignments to partition ontologies, to compute similarities between terms and to filter alignment suggestions. We test the approaches on previously developed golden standards and discuss the results.