Mr Edmond Chuc, Dr Siddeswara Guru1, Dr Anusuriya Devaraju, Dr Habacuc Flores Moreno, Dr Elisa Maria Girola
1University Of Queensland, Long Pocket,
The vocabulary development and management are crucial for the overall TERN data management strategies to describe, index and retrieve data related artefacts. A sound strategy will ensure vocabularies are adequate, accurate, up-to-date, available to their users and downstream applications in a standardised way. Without appropriate vocabularies, data may be described using inconsistent terms and representations. Consequently, this hampers machine-to-machine data search and access, which in turn may increase the cost and effort of data management with diminishing returns.
TERN uses vocabularies to describe and represent all artefacts of data and their contextual information. TERN develops vocabularies to describe artefacts such as platforms, instruments, Feature-of-Interest, Observable properties, methods, organisations and people. Vocabularies developed are two kinds including SKOS-based and instances of ontological classes. In addition, TERN also reuses existing third-party vocabularies to represent science keywords, spatial regions, spatial and temporal resolutions, UoM and species list.
In the presentation, we will describe the evolution of vocabulary development and management including the infrastructure to meet TERN and wider community needs. We will present different vocabularies developed, lesson learnt and optimum end-to-end workflow from the creation to publication. We will also describe some of the downstream applications that are using developed vocabularies.
Biography:
Edmond Chuc is a Snr Software Engineer in the TERN Data Services and Analytics Platform. Edmond’s expertise is in semantic web technologies and works in the information modelling, knowledge representation and build user-centric applications to disseminate data and related artefacts.