In this paper, we present a new semi-automatic methodology for construction of event-based ontology from the library catalogue of the largest collection in the world of metadata records of historical Hebrew manuscripts. Based on the constructed ontology, we developed and implemented a new framework for catalogue data enrichment, correction, and its systematic quantitative analysis. Finally, we demonstrate the results of the proposed large-scale analysis of three most prominent event types in the corpus, as well as a few cross-event relations and trends.
All Science Journal Classification (ASJC) codes
- Information Systems
- Language and Linguistics
- Linguistics and Language
- Computer Science Applications