Making the most of citizen science

Supplementary content information

What next for citizen science? Professor Steve Roberts on how advanced maths can make the findings of this exciting new field more useful

You must select the video player for these keys to function.

Keyboard shortcut Function
Spacebar Play/Pause when the seek bar is selected. Activate a button if a button has focus.
Play/Pause Media Key on keyboards Play / Pause.
K Pause/Play in player.
Stop Media Key on keyboards Stop.
Next Track Media Key on keyboards Moves to the next track in a playlist.
Left/Right arrow on the seek bar Seek backward/forward 5 seconds.
J Seek backward 10 seconds in player.
L Seek forward 10 seconds in player.
Home/End on the seek bar Seek to the beginning/last seconds of the video.
Up/Down arrow on the seek bar Increase/Decrease volume 5%.
Numbers 1 to 9 on the seek bar (not on the numeric pad) Seek to the 10% to 90% of the video.
 
Number 0 on the seek bar  (not on the numeric pad) Seek to the beginning of the video.
 
Number 1 or Shift+1 Move between H1 headers.
/ Go to search box.
F Activate full screen. If full screen mode is enabled, activate F again or press escape to exit full screen mode. 
C Activate closed captions and subtitles if available. To hide captions and subtitles, activate C again. 
Shift+N Move to the next video (If you are using a playlist, will go to the next video of the playlist. If not using a playlist, it will move to the next YouTube suggested video).
Shift+P Move to the previous video. Note that this shortcut only works when you are using a playlist. 

Professor Steve Roberts – Professor of Machine Learning, Department of Engineering Science, University of Oxford

There are a very large number of people who have become engaged with science through citizen science platforms. Hundreds of thousands, half a million, possibly a million people who will be online taking a few minutes a day in order to engage with these platforms and these citizen science projects. A very wide range of applications from understanding astronomical data, right the way through to amateur meteorologists who have a met station in their garden. It’s all these people that are giving up their time and their effort for free to try and help a scientific project and we really have a duty to make sure we use all that information in the best way.

Given the big interest I have in machine intelligence, I’d love to think machines were very good at doing a whole range of complex tasks; I’d be the first to admit that actually there are a lot of things that people are much much better than machines at.

Computers running the same program aren’t always going to produce almost identical answers and yet the huge range of opinions, approaches and ideas that human beings have got when they approach difficult problems, enables us to combine and tease out what we believe will be the ground truth in a much more accurate way. What we’ve been developing is a theory based on a really very elegant piece of mathematics called Bayesian Probability Theory. Bayesian Probability Theory is this under arching mathematical framework which enables us to not only work with the answers themselves, but also the underlying uncertainty which is embedded within those answers. And by doing that what we are really doing is computing in the presence of uncertainty, in a theoretical framework which guarantees, provably so, that we are going to get the best possible knowledge out of uncertain information.

One of the projects that we have been particularly involved with is trying to detect supernova, exploding stars. People are asked to try and identify whether the brightness of particular stars in that image has increased or stayed the same. If it has massively increased then it’s a really good indicator that that star might have gone supernova.

If we simply average together the scores, you get about 75 per cent accuracy. If, however, we combine them using Bayesian Probability Theory we can squeeze 98 per cent accuracy out of that same data. It really is an enormous improvement and far outshines any of the alternative approaches we can take. Where citizen science and crowd sourcing is taking us is to the understanding that we can use people, sensors and experiments which, when combined, produces something which is a lot more accurate than the single experiment.