The Natural History Open Data Challenge @ OTA16
Download as PPTX, PDF0 likes421 views
The document discusses the challenges of managing a diverse collection of over 80 million specimens at the Natural History Museum (NHM) and their ambitious digitization programme. It provides examples of data fields for specimens, references to datasets and APIs containing over 3.3 million digitized records, and potential challenges in analyzing and visualizing the collection data.
The Natural History Open Data Challenge @ OTA16
- 2. Diverse collections spanning space and time Challenge of scale: >80 million specimens! Challenge of speed (digitising within a lifetime) Ambitious digitisation programme (DCP) Institutional policy “open by default”
- 4. Higher Classification Scientific name: Thymelicus lineola (Ochsenheimer, 1808) Family: Hesperiidae Location Locality: Tilbury Docks State/province: England Country: United Kingdom Continent: Europe Decimal latitude: 51.4605 Decimal longitude: 0.3449 Collection Event Recorded by: T G. Howarth; Howarth Collection date: 31 / 07 / 1938 Most iCollections specimens will have ~30 fields containing data (over 100 different fields across all collections) There are some issues… (where is H. M. Edelsten!?)
- 6. Complete NHM Specimen Dataset (3.3M records) http://bit.ly/2goEpBB GitHub Gist – NHM API: http://bit.ly/2gtukRv iCollections Datasets http://bit.ly/2gGZub5 Even more data… http://www.gbif.org/occurrence
- 7. Potential Challenges How did collecting effort change over time? Who was the collector who collected from the most distinct localities? – can we make a ranking table and mash up data with Wikipedia or other sources? What can we learn about the collectors – who travelled the furthest or most regularly? Were most specimens collected in rural areas? Is there collection bias in particular counties? How can we make the data more attractive to difference audiences? How could we display the data in more engaging or informative ways?
- 8. Complete NHM Specimen Dataset (3.3M records) http://bit.ly/2goEpBB GitHub Gist – NHM API: http://bit.ly/2gtukRv iCollections Datasets http://bit.ly/2gGZub5 Even more data… http://www.gbif.org/occurrence
Editor's Notes
- #2: Thanks for joining us today – to my knowledge this is the Natrual History Museum’s first hackathon based on its specimen data!
- #3: 80M awesome objects - aim to get them done in under a lifetime[5:59] Default policy is openess - data and images going on the portal[5:59] Hopefully 3D at some point!