This item contains sqlite3 database snapshots, URL crawl status, and other metadata useful for doing analytics on journal OA coverage, homepage status, etc. Particularly in the context of https://fatcat.wiki. Source code: https://github.com/bnewbold/chocula
This item contains bulk metadata exported from https://fatcat.wiki. With the exception of the 'abstracts' file (for which no aggregate license or copyright claims can be made; downstream users are responsible for their use), all metadata here is licensed CC-0 (public domain release) and may be used for any purpose. Downstream users are strongly encouraged to provide attribution and link here to the snapshot, as well as give credit to upstream sources (including Crossref, ORCID, DOAJ, the ISSN...
1,381
1.4K
Sep 9, 2019
09/19
by
Internet Archive Web Group
This collection holds database snapshots (SQL) and bulk metadata exports (JSON and TSV) from https:///fatcat.wiki (an Internet Archive service)
This item contains a complete PostgreSQL SQL database snapshot from https://fatcat.wiki, in binary 'pg_dump tar mode' format. With the exception of the 'abstracts' table (for which no aggregate license or copyright claims can be made; downstream users are responsible for their use), all metadata here is licensed CC-0 (public domain release) and may be used for any purpose. Downstream users are strongly encouraged to provide attribution and link here to the snapshot, as well as give credit to...
This dump includes all tables (including oauth authentication tables which could be a privacy, but not security, concern). At this time only IA staff have accounts, so the snapshot, which is intended mostly for disaster recovery, is still public.
This item contains a complete PostgreSQL SQL database snapshot from https://fatcat.wiki, in binary 'pg_dump tar mode' format. With the exception of the 'abstracts' table (for which no aggregate license or copyright claims can be made; downstream users are responsible for their use), all metadata here is licensed CC-0 (public domain release) and may be used for any purpose. Downstream users are strongly encouraged to provide attribution and link here to the snapshot, as well as give credit to...
This item contains an example corpus of citations between scholarly documents, as extracted from the fatcat (https://fatcat.wiki) corpus as of the 2020-08-05 bulk release export. This corpus itself was generated from a fatcat-scholar "intermediate" fulltext dump which is not public, using software in the fatcat-scholar repository in mid-September 2020. See also the README for some more notes, and the "sample" file.
Contains a TSV file with SHA1, file size, wayback URLs, and metadata extracted from PDF by GROBID. Not intended for external use, but might be interested. DOES NOT CONTAIN FULLTEXT CONTENT.
This item contains some bulk research affiliation datasets from Internet Archive cataloging efforts. These are mostly strings included in research papers that indicate the institutional affiliations of specific authors (eg, with a home department, university, or company) at the time of publication. These might be useful datasets for efforts to build complete indices of research organizations, or to test normalization code that maps raw strings to organization identifiers. Attribution and links...
URL lists to PDFs on the web (and preserved in the wayback machine) which are likely to contain research materials.
URL lists to PDFs on the web (and preserved in the wayback machine) which are likely to contain research materials.
This item contains hash lists of PDF files crawled from the public web specifically to preserve the scholarly record. It does not contain hashes of *all* PDFs the archive has ever seen, only a subset. Not all of these hashes are necessarily journal articles or other research outputs, but we have reason to believe the large majority are.
Snapshot of Internet Archive (petabox) file-level metadata (eg, PDF hashes) for files under the 'journals' collection as of December 2018. Note: includes a small number of items not actually under the 'journals' collection hierarchy due to how the input item list was generated, and a small fraction (estimate 500?) of items didn't dump successfully. A bit sloppy!
This item contains datasets of homepage URLs found by hand using search engines and bibliographic metadata (eg, ISSN and journal title). The "long-tail" batch contains about 4,600 journal lookup results, with about 3,900 successful homepage URLs found. The list of journals was created in May 2020, and the lookup work completed in June 2020. IA staff member Richard Greydanus ran this batch of lookups. All of this metadata can be considered public domain, or CC-0 (Creative Commons Zero)...
Data-munged title-level metadata combined from: DOAJ, ROAD, Norwegian Register, and Internet Archive crawled metadata. See SOURCES.md for URLs of upstream metadata, and ISSN_matching.html for Jupyter notebook used to derive this dataset.
This is a mapping between: - DOIs (Crossref) - PubMed PMID and PMCID (NIH) - CORE record identifier (core.ac.uk) - Wikidata QIDs See README and scripts for details.
This is a derivative of https://archive.org/download/ia_papers_manifest_2018-01-25, which contains JSON objects that can be inserted into a fatcat catalog.
Test runs of large-scale matching algorithms (sha1 to DOI). Will likely be obsolete soon, and not useful for others.
A mirror of the Unpaywall (aka oaDOI.org) metadata corpus, primarily consisting of public open access flags for a large number of Crossref-registered DOIs (identifiers representing published journal articles and other works). For more information see: http://unpaywall.org/products/snapshot
This item contains SPARQL query exports of journal metadata from wikidata; in particular ISSN/QID mappings. The SPARQL query run is included as wikidata.sparql
About 1 million unique PDFs from Global Wayback before year 2000.
See: https://guide.fatcat.wiki/reference_graph.html License: CC-0
