BIP! Finder currently contains data for more than 138M articles and more than 1.672B citations. For each article, BIP! Finder's database contains some metadata (e.g. its DOI, author list, publication year and venue, etc.) along with its citations. These citations are used to construct the citation network of the stored scientific articles. The data used to produce the citation network on which we calculate the provided measures are gathered from the OpenCitations' COCI dataset (Dec-2022) , MAG (Dec-2021) , and Crossref (Dec-2022) . It should be noted that, we use 2nd level concepts from OpenAlex for our topics. Last but not least, since some of the indicators require for their calculation the year of the publication to be present, we keep in our database only those DOIs for which we can gather this minimum piece of information from at least one data source.
The current dataset dump containing the latest computed impact aspect scores can be found on 10.5281/zenodo.4386934 . Note that an intermediate release (between Version 3 and 4) can be found on figshare .
The BIP! NDR (No DOI Refs) dataset is also available at 10.5281/zenodo.7962019 . It contains more than 2.7M citations made by approximately 163K Computer Science conference or workshop papers that, according to DBLP, do not have a DOI.