Radio Galaxy Zoo EMU: Harnessing Citizen Science and AI to Advance Open Science Catalogues
Over the past decades, significant efforts have been devoted to developing sophisticated algorithms for automatically identifying and classifying radio sources in large surveys.
However, even the most advanced methods face challenges in recognising complex radio structures and accurately associating radio emission with their host galaxies. Leveraging data from the ASKAP telescope and the Evolutionary Map of the Universe (EMU) survey, Radio Galaxy Zoo EMU (RGZ EMU) was created to generate high-quality radio source classifications for training deep learning models and cataloging millions of radio sources in the southern sky.
By integrating novel machine learning techniques, including anomaly detection and natural language processing, our workflow actively engages citizen scientists to enhance classification accuracy. We present results from Phase I of the project and discuss how these data will contribute to improving open science catalogues like EMUCAT.
Eleni Vardoulaki, Hongming Tang, Micah Bowles, Gary Segal, Soheb Mandhai, Emma L. Alexander, Wendy Williams, Yan Luo, Lawrence Rudnick, Andrew M. Hopkins, O. Ivy Wong, Stanislav S. Shabala, the RGZ EMU collaboration
Comments: 4 pages, 2 figures, proceedings IAU397 Symposium UniversAI Exploring the Universe with Artificial Intelligence (editors Liodakis, Efthymiou)
Subjects: Astrophysics of Galaxies (astro-ph.GA); Instrumentation and Methods for Astrophysics (astro-ph.IM)
Cite as: arXiv:2509.19787 [astro-ph.GA](or arXiv:2509.19787v1 [astro-ph.GA] for this version)
https://doi.org/10.48550/arXiv.2509.19787
Focus to learn more
Submission history
From: Eleni Vardoulaki Dr
[v1] Wed, 24 Sep 2025 06:18:04 UTC (2,686 KB)
https://arxiv.org/abs/2509.19787
Astrobiology, Astronomy,
