Investigating deposits in Harvard Dataverse's Root dataverse

To prevent people from publishing spam in Harvard Dataverse, repository staff are considering publishing workflows that prevent people from publishing datasets and/or dataverses into the repository's "Root" dataverse until repository staff can verify that account creators won't deposit spam. To help estimate the number of "Root"-depositor accounts and deposits that staff may have to vet each day, the Jupyter notebook included in this dataset analyzes two years of user account data, included in the tabular file, to try to answer: How many datasets and dataverses, that have gone on to be published, have been created in Root each day? How many accounts have created more than one of these datasets in Root? How many accounts created these datasets in Root each day, and which of those accounts are first time depositors versus returning depositors? Run the notebook in Binder