A publication ontology for the CBG library

Abstract

A problem of the exponential growth of digital data collection and storage of today, is that this is not done in a standardised way, leading to inconsistencies among data sources even when the subject is the same. These inconsistencies lead to difficulties when trying to combine data from different sources which is also a problem for the CBG (dutch centre for genealogy), an organisation that helps people with genealogical research, since historical data is seldomly stored in a single database. There is a reliance on the ability to combine multiple data sources in order to provide a full picture of the story that the available data tells. The goal of this paper is to use a linked data approach to construct an ontology: a standardised structure with consistent rules about classes of entities in the data as well as relationships among them and their properties. This ontology facilitates compatibility, accessibility, and reusability of the CBG library database, providing the CBG with an opportunity to combine their data with other sources, but also allows the data to be more accessible for other institutions or research. This paper describes the step by step process of the ontology construction while taking lessons and considerations from other cultural heritage institutions that also make use of ontologies. The ontology is created by taking a subset from schema.org and using the Protégé application to turn this into an ontology. Main findings and lessons learned include the need for exact and explicit definition of the difference between series, volumes, and issues, as well as the choice of adding classes to increase the scope of the ontology versus the decision to keep it as simple and modular as possible. Linked data technologies are expanded upon step by step and the construction of a standardised ontology as done in this paper is a step towards improving usability as well as the knowledge base.