New MWP paper outlines the powerful synergy between citizens scientists, professional scientists, and machine learning
Brut uses the catalogue of bubbles identified by more 35,000 citizen scientists from the original Milky Way Project. These bubbles are used as a training set to allow Brut to discover the characteristics of bubbles in images from the Spitzer Space Telescope. This training data gives Brut the ability to identify bubbles just as well as expert astronomers!
The paper then shows how Brut can be used to re-assess the bubbles in the Milky Way Project catalog itself, and it finds that more than 10% of the objects in this catalog are really non-bubble interlopers. Furthermore, Brut is able to discover bubbles missed by previous searches too, usually ones that were hard to see because they are near bright sources.
At first it might seem that Brut removes the need for the Milky Way Project – but the ruth is exactly the opposite. This new paper demonstrates a wonderful synergy that can exist between citizen scientists, professional scientists, and machine learning. The example outlined with the Milky Way Project is that citizens can identify patterns that machines cannot detect without training, machine learning algorithms can use citizen science projects as input training sets, creating amazing new opportunities to speed-up the pace of discovery. A hybrid model of machine learning combined with crowdsourced training data from citizen scientists can not only classify large quantities of data, but also address the weakness of each approach if deployed alone.
We’re really happy with this paper, and extremely grateful to Chris Beaumont (the study’s lead author) for his insights into machine learning and the way it can be successfully applied to the Milky Way Project. We will be using a version of Brut for our upcoming analysis of the new Milky Way Project classifications. It may also have implications for other Zooniverse projects.
The Milky Way Project has now passed one million classifications since its relaunch a few months ago. The project is currently 75% complete, meaning there are still many, many images left to classify. Which is fine because in fact the project has become truly international lately – with citizen scientists around the world now able to participate in English, Spanish, German, French, Indonesian, Polish and Danish. There are more languages on the way too!
So to celebrate passing our 1,000,000 milestone I thought I’d share the homepage counter in all seven languages:
If you’re interested in helping to translate the Milky Way Project then get in touch with firstname.lastname@example.org – and there are man other translatable project too!
After a busy December and January we ran out of data a few weeks ago after 600,000+ classifications of the new images – but the wait is over! Last night a whole new, bigger, batch of data was added to the Milky Way Project. Here’s a few examples of what you might see in the data:
These new data come from the GLIMPSE 2 survey – a comprehensive survey of the middle-part of our galaxy in the infrared. We’re also going to be adding in some of the GLIMPSE 1 data (from the old version of the Milky Way Project) back into the site but with the new colour stretch. We’re doing to that to check the system works, but also because new features and structures will be visible with the change in data and colour palette.
We’re still crunching the data from the new classifications, but we’ve been able to extract lists of galaxies, EGOs and star clusters that you have found. We hope to share those with you soon.
So hop on over the milkywayproject.org and let’s add another 600,000 classifications and continue mapping the galaxy.
After a fantastic (re)launch in December and a busy January, the Milky Way Project was doing well and was about 93% complete… until about 8 hours ago. Last night, the social media powerhouse that is IFLS pointed tens of thousands of people our way and in an hour they finished the project. This is obviously great news for science but some people might be wondering what happens next.
The good news is that we have more data! The bad news is that it won’t be ready for another few weeks. In the meantime we are also working on producing some results from all your work, and you can continue to discuss things on Talk. We’ll let everyone know when we have more images to classify but for now: thank you for all your hard work and attention.
We shall return!
The Milky Way Project (MWP) is complete. It took about three years and 50,000 volunteers have trawled all our images multiple times and drawn more than 1,000,000 bubbles and several million other objects, including star clusters, green knots, and galaxies. We have produced several papers already and more are on the way. It’s been a huge success but: there’s even more data!
And so it is with glee that we announce the brand new Milky Way Project! It’s got more data, more objects to find, and it’s even more gorgeous.
The new MWP is being launched to include data from different regions of the galaxy in a new infrared wavelength combination. The new data consists of Spitzer/IRAC images from two surveys: Vela-Carina, which is essentially an extension of GLIMPSE covering Galactic longitudes 255°–295°, and GLIMPSE 3D, which extends GLIMPSE 1+2 to higher Galactic latitudes (at selected longitudes only). The images combine 3.6, 4.5, and 8.0 µm in the “classic” Spitzer/IRAC color scheme. There are roughly 40,000 images to go through.
The latest Zooniverse technology and design is being brought to bear on this big data problem. We are using our newest features to retire images with nothing in them (as determined by the volunteers of course) and to give more screen time to those parts of the galaxy where there are lots of pillars, bubbles and clusters – as well as other things. We’re marking more objects - bow shocks, pillars, EGOs - and getting rid of some older ones that either aren’t visible in the new data or weren’t as scientifically useful as we’d hoped (specifically: red fuzzies and green knots).
We’ve also upgraded to the newest version of Talk, and have kept all your original comments so you can still see the previous data and the objects that were found there. The new Milky Way Project is teeming with more galaxies, stars clusters and unknown objects than the original MWP.
It’s very exciting! There are tens of thousands of images from the Spitzer Space Telescope to look through. By telling us what you see in this infrared data, we can better understand how stars form. Dive in now and start classifying at www.milkywayproject.org - we need your help to map and measure our galaxy.
I’ve been diving into the bubbles database recently and ended up creating cutouts of all 3,744 large bubbles from the DR1 data release. From there it was an easy enough job to create this new Milky Way Project poster. It uses all 3,744 bubbles at least once (several are used more than once).
I’m currently working on three new Milky Way Project papers and will be blogging about them in the next weeks and months.
A great article from JPL’s Spitzer site, about GLIMPSE and with a lovely link to the Milky Way Project.
When drawing bubbles on the Milky Way Project (MWP) you’re looking at data from NASA’s Spitzer Space Telescope, which observes infrared light of various wavelengths from about 3 to 100 microns. Spitzer looks at warm and hot dust, as described above, and shows us where stars are forming and heating up their surroundings.
Now we have a new interface online: Clouds. When you look at clouds in our new game you’re seeing data from the Herschel Space Observatory, placed on top of Spitzer data. Herschel sees longer wavelengths than Spitzer and this means that it can detect colder material. Not long after Spitzer first began delivering science, it was noticed that there were lots of dark clouds visible in the data. These were thought to be dense, cold cores of material within the larger nebulae, where stars were still forming. Many of these Infrared Dark Clouds (IRDCs) are thought to house massive, young stars and may hold answers to some of the biggest questions in astronomy right now, such as how to massive stars form?
According at an SEO agency, when Herschel went into operation, these IRDCs were amongst the first objects to be observed and astronomers were immediately struck by an unexpected fact: lots of these IRDCs were not dense cores at all: they were simply ‘holes’ in the sky – including this striking example in Orion. Rather than looking into the dense core where stars were forming, Herschel actually began to reveal palces where one can see right through the Galaxy and out to the other side.
Doing this with computers is not accurate enough, and so to get a true catalogue of IRDCs, we’re asking volunteers to help by trying to identify them here on the Milky Way Project. If you see a bright glowing cloud then it is a true IRDC – if you see nothing, then it is a hole in the sky. Sometimes it is actually quite difficult to make out – but that’s okay, we’ll get lots of people to look at each core and take a vote.
Clouds launches today and we hope to get lots of eyes on the problem right away: visit http://www.milkywayproject.org and check it out.
It’s been two years since everyone began helping the Milky Way Project map bubbles in our galaxy (and other things too). To celebrate we’ve created another anniversary poster, featuring the names of all the participants. You can download it here (warning that’s a 19MB file) or a slightly smaller one here (5MB).
The Milky Way Project is now producing science – with two papers already published and online. You can see these and all the Zooniverse publications at http://zooniverse.org/publications. We have some new features coming to the site soon – so stay tuned.
Almost two years ago we launched the Milky Way Project and the search for bubbles in our galaxy continues at http://www.milkywayprpject.org. Today we’re pleased to to welcome a new space-based Zooniverse project into the family. The Andromeda Project (http://www.andromedaproject.org) is science in the galaxy next-door and we thought that the MWP community might like this new project. It’s very much our new sister site. We’re betting that you can help us explore some amazing Hubble Space Telescope data, to help identifying star clusters in Andromeda.