OntoLink : a system for improved ontology alignment

Ginos, Constantinos; Γκίνος, Κωνσταντίνος

Master Thesis

Author

Ginos, Constantinos

Γκίνος, Κωνσταντίνος

Date

2024-09

Abstract

Ontology Alignment is a crucial task in the field of knowledge management and semantic web, enabling the integration and interoperability of data from heterogeneous sources by matching concepts across different ontologies. Despite notable advances in this field, aligning accurately ontologies continues to present significant challenges. Many state-of-the-art systems often fall short in terms of efficiency and effectiveness, often consuming of a lot of computational power, while struggling to provide precise and reliable mapping. This thesis introduces the OntoLink system designed to address these challenges. Specifically, OntoLink uses the anonymous classes found in the input ontologies, with the help of the Manchester syntax, to discover new mappings. In addition, it employs a specific reasoning task, utilizing the HermiT reasoner, to uncover further mappings. OntoLink’s architecture has been designed to produce a set of precise output mappings, even when working with rich ontologies, that contain numerous classes and properties. Our system was rigorously tested on ontologies from the repository of BioPortal. The results of OntoLink have been compared to the mappings that BioPortal reports between ontologies. In this way, we ensured that the final results were reliable and thoroughly validated. The primary objective of this work is to introduce a newly developed system aimed at improving the efficiency and accuracy of ontology alignment tasks. OntoLink’s performance has been tested on experiments with the ontologies PSDO, STATO, NGBO, confirming that the system performs and discover mappings with high efficiency and accuracy. Overall, OntoLink shows promising results in improving ontology alignment task.

Postgraduate Studies Programme

Πληροφοριακά Συστήματα και Υπηρεσίες

Department

Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών Συστημάτων

Number of pages

Language

English

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/16954
http://dx.doi.org/10.26267/unipi_dione/4376

Collections

Τμήμα Ψηφιακών Συστημάτων

Show full item record

Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα