What's the plan?

So what exactly are we going to do with the results of this project? Here are some research proposals we would like to pursue.

Data

The first step will be of course analyzing the result. How did the users like the platform? Can we do any improvements in the future when a different survey needs the help of citizen science? One of the main objectives of this project is to prepare for the data explosion that will define astronomy in the next decade. We need to be ready to use all of the tools at our disposal to deal with the inordinate amount of objects our experiments will see, and this includes citizen science and machine learning. We need to get better at the former, and to train the latter by feeding our algorithms labeled data. With this project, you help us to do both.

On a shorter term, we plan on publishing a new set of observation of a way bigger dimension from the KIDS survey. We will do so once we have analyzed the results of this smaller scaled project and learned from all of you how to be better at our job.

Fornax population

What does the Zooniverse classification tell us about the faint inhabitants of the Fornax cluster? We will be able to explore the spatial resolution of the new fluffy objects you will find. For example, we want to study those orbiting massive galaxies to try to solve the so-called 'missing satellite' problem.

We recommend this video if you want to know more about the missing satellite problem:

Galaxy selection

One of the first things we want to do after we analyze and interpret the data you provide is to see how the Zooniverse sample compares with the commonly used selections. We would then compare the populations of existing catalogs using color, morphology, surface brightness, and compactness and analyze the difference and similarities. An interesting part is that the data you will classify has not been reduced with cuts at any point, something that is commonly done. We will study the commonly excluded objects looking for interesting properties.

Algorithms

This project also helps us better understand what the algorithms are doing 'wrong'. A lot of work has been done in the field of object identification, but with more challenging objects such as these we still need to make some improvements.

Machine learning

Machine learning is a fast growing field and an essential tool to master when dealing with big data. A labeled dataset like this can be used to train powerful object selection and identification algorithms, using models from simple classifiers to state-of-the-art transfer learning based models to more efficiently take a look at the sky.

And much more!

More than anything, we are excited about being surprised by you! What will you find? What will you notice? What unforeseen challenges will you teach us to face, and what mysteries will you uncover?

We can't wait to see!