Planet Hunters Talk

Practice, feedback

  • Transley by Transley

    Not sure if this is the right board. My question is: How do I get better without feedback? I just keep getting "nice job." I don't know if I'm missing transits or catching them, or identifying transits that aren't real. It would be nice if there were like a hundred or so stars with expert-verified transits that people could practice on and then see whether they accurately detected the transits or not. I don't know if I'm being clear, but maybe you get the idea. Thanks.


  • Choltai by Choltai

    I agree with Transley. I'm going through the same questions and wishing for the same feedback.


  • Sidler by Sidler

    The "Discussions about this star" section beside the "nice job" shows what other people are tagging that star with. This is the best feedback you want on a science project like this. Just look for a lot of tags similar to what you see in that star.


  • Planet_of_the_hope by Planet_of_the_hope

    Not sure where this questoin should go, but here it is
    Is there a way to know when the Planet_hunter organization is going to release new data, can they let us know the dates??




  • sleepygirl by sleepygirl

    I agree with the OP, more practice example with feedback would be helpful to newbies like me. There is a lot of variation in the real data, and the training examples do not really cover the trickier ones.

    Also, when I first started, it was not clear to me that i had actually "started" classifying real data - and i found the "Nice Job" pop up very confusing at first - i thought this was telling me the previously star was #pulsating, when in actual fact it was telling me to look out for this pattern in future stars... I am sure that i tagged some starts wrongly at the beginning and i've since seen other users tag stars incorrectly as well.


  • SuperGNC by SuperGNC

    A good set of training images would covering the common tags would be very nice, even if it was optional. By this, more of the help images in a library, but also a set including expert reviewed images with no transits, transits, binaries, variables, etc. These would be accessed like you would new images then you would see the expert comments and any transits marked after. This would give me a lot more clues on what to look for. That said, I'll realize I'll never ever be an expert, and perhaps from some point of view, naive (at least in my case) reviews are part of the experiment.

    I think it good to have people independently assess before seeing the comments, .