repository for retrieval of metadata

Is it possible to change the repository for metadata retreival from pdf's?

Google Scholar-retreived metadata turns out to be inaccurate. Many articles are labeled as books, in one fourth of my cases, the journal name is incorrect. And many items in the bibliography is missing. It turned out that creating citations or bibliographies from a zotero library is a lost cause.

Would it be possible to choose a well-curated repository, such as NCBI PubMed?


  • Nearly two years later and this is still an issue. Lots of information from Google Scholar that is missing incorrect, or otherwise not ready for a bibliography.
  • Google scholar works pretty well for a general audience.
    Because of its search algorithm, google is much better at "guessing" articles than other databases.
    Zotero already uses the more reliable CrossRef if it finds a doi.
    Retrieve metadata is not the main way of getting data into Zotero.

    Given all that you may see why this isn't a top priority of developers. But I'm sure everyone would appreciate progress here, so if you feel strongly about this and meant to imply that you want to volunteer for coding this, that'd be great.
  • I probably should have added that I love Zotero and think y'all are doing great work! I wish i could code, but i'm just a lowly political scientist (and a qual, at that).
  • my point was just that there are very limited resources, few people coding and already Zotero is relying (pretty succesfully I might add) on volunteers for some core functions. So you really can't expect much to happen in non-essentials (where "essentials" are defined quite fuzzily by user needs and grantee demands) unless a third party contributes it.

    I'm finding, by the way, that google scholar gets better all the time with respect to data quality.
Sign In or Register to comment.