Has there been any more work on allowing the same observation to have multiple interacting species that can each be IDed by the iNaturalist community and the observer can choose from a list of types of interaction? That would be amazing! Thanks for all the great work you do.
1 Like
Long time lurker, first-time poster (been an iNaturalist member since 2012). I am working on some smart camera machine-learning enabled classifiers on ad hoc LoRaWAN networks for remote field deployment, for a USDA Conservation Innovation Grant project, which is open source.
Boy, it sure seems like you would be the right person to talk to about making our back-end database play nicely with existing major knowledge engineering systems.
We poached a senior engineer from Microsoft, have professors from three universities, and an excellent backend database hacker from a Medical school. We want our project to play well with iNaturalist data structures. Are you available to help us understand how we can be most compatible?
2 Likes
This is not possible with an API ?
https://www.inaturalist.org/pages/api+reference
https://api.inaturalist.org/v1/docs/
https://api.inaturalist.org/v2/docs/
You do not use sounds (animals, bats, grasshoppers, mouse(birds).
I can not find the link but an open sound system (there are several) is building one up from bat sound detectors (including backround orthoptera).
There are open systems for it.
and also one for insects but all use theire own specific recognition to prevent unesccary species which do not occur on these sstems.
https://forum.inaturalist.org/t/recognize-sounds-automatically/3527/2
https://www.inaturalist.org/posts/44952-54-2-vleermuizen-vlen-avond-lezingen
https://www.ru.nl/nieuws-agenda/nieuws/vm/iwwr/2019/automatische-insectenherkenning-grip/
https://www.wildlifeinsights.org/about-wildlife-insights-ai (Google)
just for reference, iâm thinking this is the grant project: https://vetmed.tamu.edu/news/press-releases/texas-am-research-team-receives-usda-grant-to-develop-innovative-wildlife-monitoring-system/.
i sort of wonder why they want to âplay well with iNaturalist data structuresâ? iNaturalist isnât a data aggregator, but something like GBIF is. so it might be more important to understand how to play well with GBIF data structures, if theyâre intending to send that data to an aggregator. i suppose itâs possible that the intent is to send data to iNaturalist so that they can leverage the iNaturalist computer vision (and maybe community) for species classification, but that seems like a bad workflow. so then if they needed a computer vision species classification model, then maybe it would be better to train their own. and if they wanted an iNaturalist dataset for computer vision training, the AWS open data set would be the place to look for information, i think.
4 Likes
Wow, this has been around for a while. I think it would be simple enough to just draw that data from filtering certain fields (e.g. Host, Host Plant, Feeding On). In my thoughts it would be more important to keep it to unique interactions, e.g. an insect on an exclusive host, rather than a bird eating any random insect. But that it is hard to filter without numerous datapoints.
3 Likes
Wow, it would be a great feature. It occurred to me that the interactions could be simplified. Only 2 types of interaction: antagonistic and non-antagonistic. This would eliminate vocabulary problems, like oak eating a human etc.
And instead of using controlled semantics to describe relationships, symbols(arrows) can be used to visualize the flow of energy from one organism to another, as in a food chain.
Something like
Anoplotrupes stercorosus taxon page:
Interaction: Non-antagonistic
Sus scrofa-------->Anoplotrupes stercorosus
Interaction: Antagonistic
Russula xerampelina--------->Anoplotrupes stercorosus
Anoplotrupes stercorosus----------->Parasite
Anoplotrupes stercorosus----------->Predator
2 Likes
Another thread pointing to the need for a better way of recording interactions on iNaturalist: https://forum.inaturalist.org/t/observation-field-for-flowers-and-pollinators/35806/6
1 Like
Voted!
Also https://github.com/globalbioticinteractions/globalbioticinteractions/issues/822 an issue from the other side.
1 Like
With this thread being revived, my attention was caught by this sentence fragment:
I assume this was the case but is no longer. However, if there is a way (for an ordinary curator) to delete an observation field from an observation, Iâd be interested in hearing it. :-)
(Regarding the intended topic of this threadâI agree with @kuedaâs commentsâit would be useful if done well, but itâs hard to see how it could be done well.)
1 Like
Sorry for bumping an old thread but I thought folks would be interested in a prototype global feeding interaction database I recently made that basically does what a lot of folks in this thread were requesting. To be included in the database, both âpartnersâ in the interaction have to be IDed to research-grade on the Who Eats Whom iNat project.
Itâs clunky but works: https://whoeatswhom.org/
4 Likes
The data that exists on iNaturalist is already enough to derive a handful of these relations, and so they wouldnât all have to be manually added.
I was curious about this myself so I decided to work on a script in python that can derive these inter-species associations from a large iNaturalist dataset. I have an 8 GB Macbook Air with an M1 chip, and I managed to write a script that could easily handle over 800,000 observations and over 15,000 species. Hereâs a Github link for the script I wrote: https://github.com/Jacob-Deutsch-Work/Finding-Inter-species-Associations-on-Large-Citizen-Science-Datasets
Prior to this, most ecologists modeling interactions hadnât tried to systematically identify real associations from co-occurrences. Those co-occurrences (or correlations) often donât mean much because of biases in the way things are observed. On the dataset I mentioned above the strongest results were mainly gall-host plant associations. Basically this helps find a way around the problem mentioned: âthe photo only demonstrates co-occurrence and not interactionâ. As long as some users are uploading bugs and some users are uploading plants, over time that data should be enough to map out inter-species associations.
On this smaller dataset there were some artifacts alongside the known associations, so I think having some way to curate this tab and let users correct things would be good to have. Maybe something similar to how users can curate photos on a taxa. But I also think that if a script like this is run on a much larger dataset, there would be less artifacts.
I think the data available would make it possible to quantify some sort of microhabitat as well, which would be useful to display. But thatâs something Iâd have to work on in the future to prove it can be done.
5 Likes
I have noticed that interaction projects and fields are becoming more popular and it seems like every taxon group is creating their own project/fields. This is great, but it is also duplicating a lot of effort and causing a bit of confusion. I personally subscribe to https://www.inaturalist.org/projects/interactions-linked because it covers most (if not all) interactions, gets around the non-community identification issue (by linking the partner observation rather than a taxon id), and is not restricted to specific taxa. Anyway, Iâm more interested to know if there is any progress or impetus to formalise the capturing of interactions so that we donât get 5 projects and various fields added for each interaction observation.
Thanks
7 Likes