One of the great things about the algorithm is it helps newbies identify common plants. But newbies also often tend to add lots of landscape plants and such. Because the algorithm isn’t trained on these, it often gives them a wrongID of a wild species instead… For instance, LA is littered with constant observations of mountain pine species like lodgepole pine that can’t survive there (even if planted)… because the algorithm doesn’t ‘see’ or get trained on planted Canary Island Pine and other commonly planted pines… Let’s use all captive/cultivated plant observations (not sure about animals) that are the equivalent of research grade to train the algorithm also! That way the algorithm can do a better job helping newbies (and with Seek).
The vision system is trained on captive / cultivated records. However, our automated suggestions are based on vision results and nearby records, and the nearby records part of it is currently only using RG records, so if vision ranks Canary Island Pine highly, it might get knocked down by the legit lodgepole records in the Transverse Ranges.
Patrick tells me we haven’t run our accuracy tests without the RG requirement, so we’ll do that and see if it makes things better or worse.
Sort of tangential, but it’s also worth noting that our policy of assuming that cultivated plant obs don’t need more IDs means that our cultivated plant records (both images and occurences) probably have less-accurate identifications and are thus less useful for either purpose (probably why we put that RG requirement in there to begin with). I’m aware there are many who would prefer that we remove that part of our quality grade assessment, and it might help situations like this… or it might just make even more identifiers give up in the face of endless potted plants.
4 Likes
Interesting! Thanks! I guess I can close this topic then, though.
1 Like