do you want to use archived Semantic Scholar data from the API?

the checkProfiles and collectCOIs scripts in the coi repo now archive all the queries they make to Semantic Scholar (in a authors.json and papers.json file respectively). So you could use this data directly to similarly speed up reviewer assignment (and the data is much more up to date than the latest dump). I attach sample files (both are in the attached zip) from the larger profile sample csv we have been using for testing.

[authors.zip](https://github.com/acl-org/reviewer-paper-matching/files/3906258/authors.zip)

Start by reading the checkProfiles and collectCOIs scripts and inspecting the attached authors.zip, including authors.json and papers.json. Determine whether reviewer assignment can consume these archives instead of issuing the existing queries; done means an agreed implementation scope and a demonstrable speedup using the supplied sample data.

Written by the indexing model from the issue text.