Candidate Selection

Hello there planet hunters, John here again. We know that you have been anxiously awaiting word on all of the transits you have been detecting. The first batch of stars with promising transits has been released today and I wanted to give you an overview of how we selected these particular stars out of the ones you marked.

We started with the 1.2 million classifications you made between December 15th and January 16th. Any star which had a transit marked by at least 5 people and had not previously been published was our first cut. That left us with 3533 stars.

We then had a small team of astronomers here at Yale quickly go through and rate these on a 5 point scale as likely planets and eclipsing binaries. A sort of Hot-or-Not for transits. We were now down to about 800 stars that fell into one or both of those categories.

Finally, three senior dip spotters went more carefully through this list, rating them again. Any star which was marked as either a possible planet or eclipsing binary with a score of 4 or better made it onto this first list of candidates. 90 possible planets and 42 possible eclipsing binaries!

There were many exciting transits that did not make the cut. Mostly it was because we need more data. You will notice that there are some single transits in the list, but there were just so many good ones it was hard to leave them out.

Our next step is to model these transits and weed out any more that may look promising by eye but aren’t quite as regular as they appear. This will also allow us to add radius and period information for most of the stars. Additionally, we will be including all of those stars where you identified existing planets, planet candidates, and eclipsing binaries from published works. I can already tell you that you easily found all of the published confirmed Kepler planets which were in the data.

So, head on over to the Candidates pages, or check out the two new links on your profile page which shows you any planet or eclipsing binary candidates which you marked a transit on.

Thanks for all of the hard work!
John M. Brewer


About The Zooniverse

Online citizen science projects. The Zooniverse is doing real science online,.

9 responses to “Candidate Selection”

  1. Andres Eloy says :

    Hi team of Kepler mission, I very excited with the results of the search for exoplanets, I found 7 and a binary star.

    Thanks for this great opportunity!

  2. kianjin says :

    I have a question regarding EBs. If a suspected EB was spotted, but the transit button wasn’t clicked, how would this flag the star to you folks to consider?

    For example, with some very short-period pulsators with the ‘shutter-effect’ – it’s not possible to mark any transits. If we go to the talk page and put in a hashtag – would that bring it to your attention? And if so, which hashtags?

    I think many of us have found quite a few of these unlisted EBs, at least not in Prsa’s Kepler EB list. (Now I found out that Coughlin et al has yet another list that needs to be consulted as well)

    Just so happens a few hours ago I found this possible ELV, not in Prsa or Coughlin: KID 6700870 at, nobody else commented on it, perhaps bypassed as just a pulsating variable, but periodogram brings out the sinusoidal waves, and its period while long at 1.18d for an ELV, is shorter than some in Prsa’s list.

    Does a star need 5 ‘classify’ visits to get it flagged? If we put a star in a collection and someone comes along to comment, would that count? (I guess not) What about stars that we didn’t classify, but while perusing in someone else’s list, we spot something interesting, like hidden transits?

  3. John M. Brewer says :

    I realized that I left one very important statistic off of my blog post. 719 users had a hand in selecting our candidates. This is incredible given that we only had 132 candidates and really proves the power of citizen science.

    This is not a game solely for a few chosen experts, but one where our collective intelligence is brought to bare on a difficult problem. The experts can definitely help, but it is the group which is making the discoveries.

    Congratulations to all.

  4. michaelshopkins says :

    Why is the “Candidates” page one that requires a login? This is a page that should be made easy to access by anyone including those who are not looking though light curves.

  5. cosmicphil says :

    Hy everybody,

    I’ve etudiate less than 1100 light curve, for one exoplanet candidate and one binary star system candidate, on this first extraction.
    I’m so full of lots of emotions … it’s quiet like one of my child dream was come true !
    Thank’s for all of you who have made my dream come true.
    A question : are our discoveries in the 1235 Kepler new worlds candidates i’ve seen on the press conferens this day ?

    From Britany, France
    Philippe MICHEL

  6. cosmicphil says :

    Hy ; for my candidate, i’ve the answer to my question : i’ve seen all new Kepler team candidates (not all, but just search for the id kepler star number in the pdf on Kepler website), and it’s not in the new list … what exciting and amazing is this quest of new worlds, better than everything i did before.
    So thank you for all of you ones again !

    Always from Britany, in France, in Europe, on our beautifull Earth …

  7. nodders says :

    Sorry for being a bit dim, but at what stage am I asked to classify a star as a “eclipsing binary”? Or is it just that some “transits” are classified by experts as “eclipsing binaries”?

    None of my stars are in either list, but I did pick up a simulated transit.

    I am probably someone with lots of false positives. I hope that erring on the side of caution does not introduce bias in your analysis.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: