OpenAlex Integration into Unsub

[Scott Chamberlain]

Hi all! Happy almost weekend. 

We just released today a new version of Unsub integrating data from OpenAlex (https://docs.openalex.org/).  

For now the integration of OpenAlex is limited to journal metadata (e.g., ISSN, title, publisher, etc.) as well as journal concepts/subjects (e.g., Medicine, Ecology). Later we'll integrate additional OpenAlex data. 

To be clear, subjects throughout Unsub (in the user interface and in data exports) are now from OpenAlex. 

Scenario exports used to have two subject fields (subject, era_subjects (from Excellence in Research Australia)), but now have three subject fields: subject, subject_top_three, and subjects_all. See the docs (https://docs.unsub.org/reference/data-export) for further details about these columns. In addition, all publishers have the subject column populated in data exports now, whereas only Elsevier dashboards had that column populated previously. You can find out more about a concept by navigating in your browser to https://openalex.org/ + the concept id (e.g., C2778407487), for example, https://openalex.org/C2778407487 . The OpenAlex concept Ids are located in the subjects_all column of your data export. 

Note that we've been in touch with Eric that runs Unsub Extender - and the new data exports should work with Extender now or soon. 

This change of journal metadata does affect what titles are included in your Unsub dashboards because of improved journal metadata in OpenAlex. To quantify the change, we analyzed the same 212 randomly selected scenarios before OpenAlex and after integrating OpenAlex, and there are very small differences (percentages presented because absolute numbers are not meaningful when there are vastly different numbers of titles among scenarios)

• On average number of titles included in forecasts differed by 0.62% (max 2%)
• On average forecasted costs differed by 0.18% (max 5%)
• On average forecasted access (fulfillment) differed by 0.05% (max 0.4%)

A subset of these differences result from journal metadata corrections that can be made, and we are working on fixing those as quickly as possible. 

Associated Github release notes: https://github.com/ourresearch/jump-api/releases/tag/v1.7 

Please let me know if you have any questions about this change at sup...@unsub.org

Best, Scott