Hamilton- Interactive Ontology Learning
TL;DR
Hamilton is presented, an interactive application for ontology learning from relational databases that combines automated modeling suggestions with expert-driven refinement and illustrates how automatic modeling and human expertise can be effectively combined to construct high-quality ontologies from relational databases.
Abstract
Relational databases store large amounts of structured data, but their schemas often do not directly correspond to semantic data models such as ontologies. Transforming database schemas into ontologies is therefore an important step for enabling ontology-based data access and data integration. However, this task is challenging because relational schemas are designed for data organization and integrity, not for representing domain semantics, and modeling may be context-dependent. In this demonstration, we present Hamilton, an interactive application for ontology learning from relational databases that combines automated modeling suggestions with expert-driven refinement. The demonstration showcases the full ontology learning workflow and illustrates how automatic modeling and human expertise can be effectively combined to construct high-quality ontologies from relational databases.