Closed diatomsRcool closed 2 years ago
It happens because these words are also valid names.
For example https://verifier.globalnames.org/api/v1/verifications/Tsukada
The same problem happens with some geographical entities like America
for example.
To avoid false positives we need to study contexts and apply machine learning techniques I guess.
Added https://github.com/gnames/gnfinder/issues/62 for name finding algorithm. If I add the names to grey dictionary, it should help to hide such stand-alone names from results.
More names (I can provide the data package number if needed)