Complex alignment of modular ontologies

Abstract

Ontology alignment, a critical process in the Semantic Web for detecting relationships between different ontologies, has traditionally focused on identifying so-called ”simple” 1-to-1 relationships through class labels and properties comparison. The more practically useful exploration of more complex alignments remains a hard problem to automate, and as such is largely under-explored, i.e., in application practice it is usually done manually by ontology and domain experts. In this research, we explore the current state-of-the-art techniques in ontology alignment, focusing on several key areas. This project explores key challenges in ontology alignment, focusing on improving accuracy through enriched information, understanding performance differences between synthetic and real-world data, and evaluating how question formats and data representation affect user interaction and alignment outcomes. Recently, the surge in Natural Language Processing (NLP) capabilities, driven by advancements in Large Language Models (LLMs), presents new opportunities for enhancing ontology engineering practices, including ontology alignment tasks. This research explores the application of advanced language models to address complex challenges in knowledge representation and alignment. By adopting a prompt-driven methodology and incorporating rich, context-specific content modules, this study presents a novel framework aimed at enhancing automation in alignment processes. The proposed approach contributes to advancing intelligent systems capable of improving accuracy, scalability, and efficiency in knowledge integration tasks, with broad applicability across diverse domains such as information management, business intelligence, and decision support systems