charvolant
commented
1 year ago The two records are https://biocache.ala.org.au/occurrences/bb9229bb-4d5a-41fc-8357-8f6bf261ce3c and https://biocache.ala.org.au/occurrences/959ebf16-e7d2-4f28-bcb0-8baf409d74e7 name supplied as Grevillia striata (note -ia ending) without additional taxonomic information.
20230329-2 index
-
Grevillia striata correctly matches https://id.biodiversity.org.au/node/apni/2911702
-
Grevillia striata matches https://biodiversity.org.au/afd/taxa/a9f5f4c8-62f1-41c0-bc1d-f6491592a966 with a higher order match
-
The fuzzy search for the misspelled name is
+genus_ex:grivila +specific_ex:strata +infra_ex:<null> genus:grevillia (is_synonym:F)^20.0with taxamatch treating the -ea and -ia endings as different. Grevillea striata -> GRIVILIA STRATA and Grevillia striata -> GRIVILA STRATA This also makes ala-name-matching-2 resort to a higher order match, since the misspelling is different enough to not be accepted.- This would require a retooling of the taxamatch soundex, likely to lead to new, different and interesting sorts of errors.
-
I think the only possible fix for a problem like this is to include a specific orthographic invariant in dr7933
The ALA has mismatched several records of Grevillea striata R.Br. to the beetle genus Grevillia Bordoni, 2005. The two Ausplot records have a misspelling of Grevillia, but instead of arriving at a probable misspelling of the combination, Grevillea striata, the names matching has ignored the specific name and matched on genus.