Machine Readable Semantics in a Data Science Workflow

Abstract: FAIR data principles (findable, accessible, interoperable, reusable) have been well received in the data science community. In recent years, many best practices of open data have demonstrated their value to facilitate data-driven discoveries in scientific research. Among those studies, semantics remains as a key topic of wide interest, which is relevant to terminology, data models, formats, metadata, ontology, vocabulary, knowledge graph, and many other subjects. In particular, with the thriving of Web-based data sharing and discovery activities and the extension of FAIR principles to data analysis software and other objects in open science, it is worth to have a reflection on the role of semantics in the FAIR principles and make recommendations for future works. In this presentation, we will review a few recent projects that have worked on semantics of data and implemented it in different steps of the data science workflow. We will analyze the pros and cons of the current practices, and will also present a vision for potential future improvements.
Speaker(s): Xiaogang (Marshall) Ma, Ph.D.,
Agenda:
Social event at 5:30 PM, Local only.
Techical presentation and Q & A session 6 to 7 PM. Local and online.
Room: JEB thinkTANK (First Floor), Bldg: Janssen Engineering Building (JEB), University of Idaho – Moscow Campus, Moscow, Idaho, United States, 83844, Virtual: https://events.vtools.ieee.org/m/327001