Semantic Modelling of Ancient Sources Using Ontologies

This is the authors' abstract. We don't add key points for this paper.

Ancient literary evidence is commonly dispersed across works, editions, and citation systems, making passage-level thematic retrieval and aggregation difficult. This study presents SourceArchiveOnto, a lightweight application ontology for representing ancient literary works and their passages together with authorship, citation structure, subject, genre, chronology, and selected geographic metadata. The ontology was developed from explicit requirements and competency questions and reuses established Semantic Web vocabularies where their published semantics match the intended relations; locally defined terms are used where no suitable reusable term was identified. A populated proof-of-concept dataset concerning Western Locris demonstrates the model’s use rather than attempting comprehensive coverage of the region or of ancient literature. Evaluation combines dataset statistics, competency-question-driven SPARQL queries, logical consistency checking, and inspection of representative instances. The results show that the schema supports reproducible passage-level retrieval and corpus description across the represented works. The evaluation does not establish cross-regional generalizability, large-scale performance, or comprehensive modelling of provenance, manuscript transmission, and competing interpretations; these remain subjects for future extensions and external validation.