/aif-fep-db

A database of publications related to active inference and the free energy principle

Primary LanguagePython

aif-fep-db

A database of publications related to active inference and the free energy principle.

Features

  • Helper functions for filtering and sorting.
  • Removal/cleaning functions for papers that should not be in the database.
  • Custom tags to add to the database.

Downloading the database

Examples

See how to add the database and add your own tags. See how to cite and download papers.

Possible future features

  • squlite3 support as an alternative to CSV database

TODO and Features

  • Separate tags file + db. In other words, the db stays intact always. Separate file is created.
  • Tag file is now just a metadata file.
  • Removing papers is in the tag file will the remove tag or maybe a separate section
  • Adding papers should happen in a separate way. One way is the through auto-download from pubmed and Arxiv. The other is manually adding for papers that are found in neither place.
  • Support for books or book chapters and masters theses/disserations
  • Summary field (i.e. even shorter abstract.)
  • Download PDF functionality (Paperpile or otherwise?). Maybe from my current repository of PDFs. GoogleDrive request?
  • Interactive add tag mode. Goes through the table one at a time in terminal, displaying the row. Lists all current tags. The user can then add tags. Accepted commands include "add tag [tag]" and "next file" or "previous file". Only lists files without tags yet.
  • ArXiv and BioArXiv support for preprints or open-access material
  • Bib-tex export and or citation support. Perhaps option for APA as well.
  • Links to all open access papers.
  • Get abstract method?
  • Switch to UUID instead of pmid to account for ArXiv papers, preprints, and open access papers
  • DOI links
  • Layout needs pages for about/info, tag reference, minimal data stats, and perhaps even a section on possible queries of interest (like early AIF papers)
  • SQL support or some way to search the table other than just filtering by clicking the columns
  • Work on auto-refresh. Should check the database to ensure that the paper is not already there. Then it adds to the database. Then the website will auto-refresh and add papers every week or so.
  • Eventually expansion to an ontology of tags
  • Public and private modes for editing the table or custom tags
  • timestamp on website to show when data was last download from pubmed. How recent the db is.
  • Small ArXiv paper to cite later. Put this in the about section.
  • Preloaded libraries for Paperpile, Zotoro, or Mendeley?
  • Network analysis of authors and labs.