11:05 〜 11:20
[U06-07] Using semantic technologies to facilitate open science: our experience with TickBase and Mindat
キーワード:Semantics, Open Data, FAIR principles, Data Science Workflow
Semantic technologies such as vocabularies, ontologies, knowledge graphs, and large language models have gained rapid adoption in geosciences. Over the past decades, geoinformatics researchers have leveraged these technologies to enhance data flow and scientific discovery. However, newcomers may find the technical terminology overwhelming and struggle to grasp the practical benefits. This presentation will take an empirical approach to quickly illustrate use cases of semantic technologies aligned with a typical data science workflow, from data collection and preprocessing to advanced analytics, data products, and result communication. Specific details will be given to two recent projects: TickBase and Mindat. The former has deployed LLM in data cleansing and semantic embedding in data search. The latter has leveraged community standards in its data structure and persistent identifiers in its data service. Those technical developments aligned with FAIR (Findable, Accessible, Interoperable, and Reusable) data principles, and the resulting open data services have underpinned meaningful and trustworthy of scientific workflows and results.