Over the weekend we implemented a function to automatically enforce rules for the consistent (re-)naming of places. E.g. "Paris-Frankreich" will be automatically rewritten to "Paris (Frankreich)" if a user enters the name while working in German. This way it is easy to identify as a duplicate and merge it with the existing entry "Paris".

These rewrites are applied both in musdb and nodac. See more in the blog: https://blog.museum-digital.org/2023/11/27/automatically-enforcing-consistent-naming-of-places/

I think that it's especially Ukrainian entries that hold a lot of potential for automatic cleaning. Take for example "Україна, Запорізька обл., м. Запоріжжя, Приморський р-н": We could automatically detect the oblast, м. and р-н and rewrite it to "Приморський (Запоріжжя)" or something similar. I'd need help with the meaning of the different abbreviations however, and with identifying the different logics by which such entries are written. And also with an idea how the names should look like for the easiest detection of duplicates and enrichment without accidentally leading to the erroneous merging of actually different places.

It would be great if we could collect some examples in the form of "from" > "to" relations.