- PI: Kyle Cranmer
- Engineer: Sinclert Pérez
- Data Scientist: Quynh M. Nguyen
It is well known that terminology and jargon are one of the barriers to communication across disciplines. The same mathematical concepts have different names (eg. ELBO and free energy).
Similarly, dialects vary geographically. Sometimes geographical maps that show how different words are used for the same concept are very helpful. Dialect map proposes to make a similar map, where the geographic map is replaced with a 2-D visualization (embedding) of literature and then overlays how a certain concept is named on this fixed 2-D representation of the domain literature.
This would be done with a hand created list of jargon terms, displayed in an interactive website.
A more complete description of the project could be found on NYU Data Science blog.
- Project UI: the public web interface.
- Project Data: the community sourced jargons.
For details about the software deployment and interconnection, check out the architecture docs.
- Similar project: https://blog.paperscape.org
- Front code: https://github.com/paperscape/paperscape-mapclient
- Back code: https://github.com/paperscape/paperscape-backend