What is the Community Taxon?

The Community Taxon (or Community Identification) represents what taxon the iNaturalist community thinks is depicted in an observation. In general, we try to choose a taxon that more than 2/3 of the identifiers agree with. Each observation with at least two active identifications has a Community Taxon.

Sometimes this means choosing a higher-level taxon that contains a number of disagreeing taxa (e.g. you think it's a kingsnake and I think it's rattlesnake, so iNat chooses suborder Serpentes which contains all snakes). The algorithm also slightly favors dissent, because we've found that dissenters are often correct.

More Details

A Research Grade observation must have (among other criteria) a Community Taxon. If an observation has only one identification, it won’t have a community taxon. All observations with at least one identification will also have an Observation Taxon. The Observation Taxon is the taxon iNaturalist uses when sharing observations with data partners, linking observations of the same taxon on the site, updating your life list, etc. In most cases the Observation Taxon will eventually be set to the Community Taxon, but sometimes they will differ, especially before the community has settled on an identification. For example, if you think it's a snake (suborder Serpentes) and I think it’s a kingsnake (genus Lampropeltis) the observation taxon will be at kingsnake (supported by my identification only) but the Community Taxon will be at Serpentes (supported by at least two identifications).

If for some reason you don't agree with the Community Taxon, you can reject it on your own observations, which means that Observation Taxon will never be set to the Community Taxon and will always match your own ID. It also means your observation can only become Research Grade when the community agrees with you. If you don't like the whole idea of the Community Taxon, you can opt out of it entirely by editing your settings.

The Algorithm

For all identified taxa and the taxa that contain them (e.g. genus Homo contains Homo sapiens), score each as the ratio between the number of "agreements" (cumulative IDs for that taxon) over the sum of the cumulative IDs, "disagreements" (the number of IDs that are completely different, i.e. IDs of taxa that do not contain the taxon being scored), and "ancestor disagreements" (the number of more conservative IDs that explicitly disagree with the finer taxon being scored). For the identified taxa that have a score over 2/3 and at least 2 identifications, choose the finest-ranked taxon (e.g. if genus Homo and species Homo sapiens both have scores over 2/3 and both have 2+ identifications, choose Homo sapiens because it has a finer rank).

score = cumulative count / (cumulative count + disagreement count + ancestor disagreements)

Identification count = # of identifications for an individual taxon

Cumulative count = # of identifications for an individual taxon and all its descendants

Disagreement count = # of identified taxa that are not among a taxon's ancestors

Ancestor disagreements = # of identified taxa that are among a taxon's ancestors, but disagree with the taxon (i.e. "I think it is in the genus but I disagree it is that species") 

On any observation with a Community Taxon, you can see how the observation's identifications affect the Community Taxon algorithm by clicking on "What's this?" or "About" on the observation's page on the iNaturalist website.