If #377 or #222 are integrated, and also when manually canonicalizing, it would be helpful to have an admin view that highlights possibly incorrectly canonicalized works.
One possibility could be to use trigram similarity measurement using the pg_trgrm module, or just by simply calculating the Levenshtein distance of each nomination's proposed work name from the work name it's associated with.
In either case, we'd want to render this into a table for the work, showing possibly mis-canonicalized nominations at the top or otherwise highlighted, and provide a control to de-canonicalize them.
If #377 or #222 are integrated, and also when manually canonicalizing, it would be helpful to have an admin view that highlights possibly incorrectly canonicalized works.
One possibility could be to use trigram similarity measurement using the
pg_trgrmmodule, or just by simply calculating the Levenshtein distance of each nomination's proposed work name from the work name it's associated with.In either case, we'd want to render this into a table for the work, showing possibly mis-canonicalized nominations at the top or otherwise highlighted, and provide a control to de-canonicalize them.