Giving Zooniverse a Face: A Citizen Science Facebook Application

Southern Illinois University At Edwardsville, Edwardsville IL

Investigators

Abstract

The Internet is systematically changing how professional scientists and the public engage with astronomical data and with one another. One of the most striking examples of new ways astronomers and the public are collaborating is the "Galaxy Zoo." This online citizen science project has attracted over 230,000 people from around the globe to make over 100 million classifications of the morphology of galaxies. These classifications, taken in aggregate, have been shown to be as accurate as the work of professional galaxy classifiers and have led to the publication or submission of 14 papers on topics as diverse as galaxy mergers, active galactic nuclei behavior, and machine learning. The Galaxy Zoo is now transforming into a project called Zooniverse that will provide a framework for a diverse set of citizen science tasks. As the Zooniverse grows, we need to find effective ways to expand the Galaxy Zoo user base, and this project uses Facebook to address this fundamental need. One of the fastest growing virtual communities in the world, Facebook potentially offers exceptional leverage to scale up the number of research questions that can quickly be answered in the realms of both astronomy and education. This project will develop the first two Facebook applications to engage citizen scientists. The first will focus on getting people to do science by inviting them to pick from a "Rogue's Gallery" the galaxy that best matches a "Most Wanted" galaxy. The second will use social networking to build the user base of the main Zooniverse website by linking users' Facebook and Galaxy Zoo profiles, allowing users to share socially the results of their citizen science work, to play trivia games, and to give friends "galaxies" to build their own "clusters". The results of these galaxy classifications will be used in a Principal Component Analysis looking for new ways of categorizing galaxies. The sample of roughly 250,000 galaxies provides a valid space in which to define meaningful categories of galaxy morphology, deriving for the first time natural classification classes from the appearance of a large sample of galaxies, instead of sorting them into categories defined a priori. This project will also pave the way for future applications to migrate into Facebook by defining what characteristics do and do not attract users to engage in citizen science online. This will allow other projects to engage users in astronomical research and will inform those working on non-astronomy citizen science about how better to create future applications.