A Novel Model for the Semantic Enrichment of an Ontology Alignment System

Abstract

In many applications, ontology alignment is a difficult challenge, and it is a major concern for interoperability and domain specialists in information systems. Although there are various methods for ontology alignment, most of them ignore ontological links such as subsumption and aggregation. Furthermore, traditional alignment methods provide little or no information about the underlying structure of the correspondences between ideas that they identify, limiting them to basic links between matched concepts. However, many actions, such as ontology mergers, ontology evolution, or data conversion, require more detailed information, such as the actual relationship type of correspondences matches or information about the cardinality of a correspondence (one-to-one or one-to-many). We employed an enrichment technique to build an upgrade to the present ontology alignment tool (Falcon AO++) that recognizes and adds more semantic information to a created ontology mapping in this study. The improved mapping now supports equal, is-a, inverse is-a, part-of, has-a, and related semantic connection types. To enable semi-automated mapping to correct and detect more sorts of correspondences, the enrichment technique leverages a variety of linguistic, structural, and background knowledge. As an outcome, more expressive mappings may be created. In terms of precision, recall, and f-measure, Falcon-AO++ and eFalcon-AO perform better by 18.1 percent, 20.2 percent, and 18.6 percent, respectively.