Weird Galaxies and Wonderful Animals: Why Big Data Needs Crowdsourcing to Find the Unusual in the Universe
Prof. Lucy Fortson
School of Physics and Astronomy
University of Minnesota
What do galaxies, lions, Higgs bosons and ancient texts from the Cairo Geniza have in common? Each of these research areas suffers from a similar problem: too much complex data for researchers to properly analyze. To make progress, researchers have turned to the general public to ask for help. Zooniverse is the largest online citizen science platform in the world, with nearly 3 million participants on over 450 projects performing tasks like classification or marking, on images from camera traps in the Serengeti to those from Astronomical Sky Surveys. Within astronomy alone, serendipitous discoveries with Zooniverse include new categories of galaxies and extra-terrestrial planets. But new generations of instruments are creating ever-larger numbers of images and other data that need to be classified. Machine learning can now do a lot of the tasks that humans can do – and they can do those tasks more efficiently. But can machines make serendipitous discoveries? In this talk, I will take you on a quick tour of the engaging projects in the Zooniverse – from the Lions in the Serengeti to galaxies in the furthest reaches of time and space. Along the way, I will describe the issues that researchers now face with “Big Data” in detecting scientifically interesting anomalies, and how harnessing crowdsourcing techniques to optimize the combination of human intelligence with artificial intelligence is revolutionizing how science is being done.
Hosted by Prof. Wayne