Snapshot Serengeti Talk

Opps, error

  • Veegee by Veegee

    I just noticed a human chopping grass and clicked what I thot was the "human" button, but it was actually
    the zebra button, and I noticed it just as I clicked submit....so, if you wonder why a human was classified as a zebra,
    that is why lol

    Posted

  • kosmala by kosmala scientist

    No problem, Veegee. We expect some mistakes like this and so show each image to lots of people. So don't worry about it.

    Posted

  • ci.el.o by ci.el.o

    i just click finish by mistake, and didn't report a large herd of elephants. is there a way of going back and fixing it. ?

    Posted

  • dms246 by dms246 moderator in response to ci.el.o's comment.

    No, once you click on Finish, that's it. Don't worry though - as @kosmala says above, occasional mistakes like that are expected - after all, we're all human, and can expect to make mistakes now and then. 😃 When all the individual classifications for an image are pulled together, the occasional mistake won't significantly affect the data analysis.

    Posted

  • xiqiqi by xiqiqi

    how do i say 'i don't know'? i classified a moving entaglement as 'nothing there' because i did not know what else to say.

    Posted

  • dms246 by dms246 moderator in response to xiqiqi's comment.

    What the scientists want in cases like that is your best guess, xiqiqi. I know it feels "wrong" to simply guess, but in fact, when your best guess is set alongside all the best guesses of the others who have been shown the same photo, it can actually provide quite a lot of information. There's a sticky thread that explains how it works, and which contains a link to a blog post by @kosmala which goes into more detail about how they analyse and extract information from the data:

    Why don't we have an "Unidentifiable" option when classifying photos?

    Posted

  • chaoticidealism by chaoticidealism

    So how many people see each photo? A dozen? As many as you can get to look at them? How many people's guesses does it take to get a good idea of what's in the picture, anyway?

    I bet there's a math nerd out there who has figured out the statistics of it, and now I can't help but be curious, because, alas, I am a math nerd myself...

    Posted

  • areinders by areinders moderator in response to chaoticidealism's comment.

    If you read the blog connected with this project, Margaret explains a lot about how they analyze the data. It really very interesting.

    Posted

  • chaoticidealism by chaoticidealism

    Thanks! Interesting indeed.

    Posted