Read Time:1 Minute, 10 Second
Relevant notes from a Metadata: Description and Access guest lecture by Corey Harper of Elsevier
- Unlike Carnegie Arttracks team, not skeptical of linked data as a tool in implementing machine learning.
- “Ease of integration across data sources – merging graphs”
- ETL – Extract, Transform, Load
- Formal definition of how data elements relate to one another
- 2009 Tim BL TedTalk – linked data publication exploded
- 2011 W3C LLD Incubator – Encouraged experimental linked data projects
- “Don’t just publish data, think about how people are actually going to use it”
- A bunch of nodes is meaningless to most people
- A Thinkbase/Freebase browsable interface
- Google Knowledge Graph – the little info boxes you see when you search
- They bought Freebase, hence why it disappeared
- No one knows or cares that it’s linked data
- RelFinder – http://www.visualdataweb.org/relfinder.php
- Can search for associations between two entities across sources.
- Linked Jazz-style
- SNAC – “Facebook for dead people”
- Prosopography – the study of an aggregate group based on systematic biographical study of individuals
- Data in Narrative
- Trove – https://en.wikipedia.org/wiki/Trove (Australian Europeana/DPLA)
- http://discontents.com.au/
- Open Context
- Pleiades
- Books:
- The Second Machine Age
- Weapons of Math Destruction
- Argument for linked data in libraries – this data may be less biased than data coming from Google/other corporate source