Duplicate Handling

I just used the duplicate processing for the first time. The UI is great! It worked really well. Very easy to use.

A handful of pairs came up as possible dups but weren't. The volumes, issues, pages and dates were different so it's puzzling why they appeared.

A bunch of duplicates in the database were not found by the processing, apparently because the authors' first names were spelled out in one case and just the initials appeared in the other. It seems that this could be easily fixed.

Duplicates were also not found if the title ended with a period in one case and not in the other.

Other cases that were missed are not so obvious. Could a means of manually picking pairs to be merged be added for these cases? Currently the pair must be updated so the duplicate algorithm finds them and only then you can merge them. Or am I missing something?

Again, I'd like to congratulate the creator of the UI on a nice job. Duplicate processing is a great addition to zotero.
  • The detection algorithm can definitely still be refined. (It'd actually be great if people contributed patches for that on zotero-dev. That code is pretty self-contained.)
    Could a means of manually picking pairs to be merged be added for these cases?
    As long as the items are of the same item type, you can. Right-click on the items not in duplicates view and select Merge Items.
Sign In or Register to comment.