W3C presentation on CoronaWhy Knowledge Graph

CoronaWhy Knowledge Graph presented by Slava Tykhonov (DANS-KNAW, the Netherlands) for the SEMANTIC WEB IN HEALTH CARE AND LIFE SCIENCES COMMUNITY GROUP (https://www.w3.org/community/hclscg/)

Clickable slides

Session 32
April 24, 2020
Session 31
April 23, 2020
Summary:
  • 2:32 - Zooming into teams so that everyone can understand their functionality
  • 5:39 - Daily call is just a glimpse
  • 7:34 - Our webinar and inviting medical experts
  • 8:36 - The Converger’s call
  • 9:26 - Team Reporting: Risk Factors (@Mayya)
  • 12:07 - Software Infrastructure Channel
  • 15:19 - Team Reporting: Task Transmission (@Christine Chen)
  • 18:14 - Team Reporting: Task Geo (@Manuel Alvarez)
  • 20:18 - Task Geo’s team structure and next steps
  • 23:00 - Splitting teams into fractals (ex. VT and future Geo)
  • 28:34 - What is task-geo’s common theme? Data Infrastructure
  • 32:48 - Team Reporting: Vaccines/Therapeutics (@Shannon Cahill-Weisser on behalf of @Dan Sosa)
  • 33:57 - We need medical experts
Session 30
April 21, 2020
Summary
  • 01:12 - Artur's health and wellbeing and notice of day off.
  • 01:32 - Mental health check-in's.
  • 02:10 - First Podcast Episode is out on platforms https://open.spotify.com/show/3coFCjB....
  • 02:20 - Kaltura sponsorship platform for future video calls, communications team are going to test it out first.
  • 02:50 - Post-Submission Webinar planning progress and the intended audience.
  • 03:35 - Ansun's made a Zapier Slack link collector for document storage.
  • 04:30 - Pets talk, maybe a Pets Slack Channel.
  • 04:58 - #task-risk - Knowledge Graph needs and future approach including referencing what the search team are working on. No blockers.
  • 07:27 - #task-TIES - Still developing ideas and are making plans but are thinking about validation long term of data they are testing from. intend to have projects laid out by Friday. No Blockers
  • 10:54  - #task-VT - Formulated a set of potential projects, no blockers.
  • 11:46 - No one from #task-datasets
Session 29
April 20, 2020
Summary
  • 0:00 - The study of the CoronaWhy ant colony
  • 3:00 - Direction: Post-Kaggle submission
  • 7:20 - The product that we are building (AI literature review)
  • 8:38 - Understand the ontology of the actual data in CORD-19
  • 12:05 - The importance of data enrichment and preprocessing
  • 13:28 - A Round 1 Assessment
  • 14:09 - More people can make a better infrastructure
  • 15:52 - The Kaggle CORD-19 dataset
  • 18:48 - Where and how do people engage?
  • 21:12 - Can we publish a scientific paper of our research?
  • 25:23 - How do the vertical teams proceed?
  • 28:01 - The Search process: Integrating vertical teams with the NLP search engine team
  • 31:28 - Structuring of the teams to prepare for Round 2
  • 33:13 - Andrea’s anonymous roster (matching skills with tasks)
  • 36:36 - Team Reporting: Vaccines/Therapeutics (@Dan Sosa)
  • 37:38 - Team Reporting: Task Transmission (@Christine Chen)
  • 40:00 - Team Reporting: Risk Factors (@Mayya)
  • 43:56 - Team Reporting: Task Geo (@Manuel Alvarez)
  • 46:16 - A tighter schedule for the general call
Session 28
April 18, 2020
Summary
  • 3:29 - Progress with the search engine
  • 8:02 - Many different/diverse arenas of study
  • 8:48 - Presenting our results to the medical community (through webinar)
  • 11:00 - What is our final product?
  • 15:00 - Our work compared to the creation of Google (a science experiment)
  • 18:19 - Selective search based on the audience
  • 20:35 - A Human input driven system
  • 22:08 - How to crowdsource our need for human input
  • 26:11 - We need a solid structure (outline)
  • 29:52 - Picture CoronaWhy as a guild
  • 34:51 - The benefits of CoronaWhy gamification
  • 37:00 - Guild Setup: Getting Skills Information from People
  • 40:57 - The Social Aspect of CoronaWhy
  • 45:03 - Interview with AI2
  • 46:29 - Potential to impact policy
  • 50:54 - Participatory Economics Channel
Action Items
  1. Work on interview document with AI2