Outliers within an astronomical catalog hold the potential to offer new insights into astrophysical phenomena.  Machine learning is used to identify outliers in the Chandra Source Catalog v2.0. 

Samantha Watkins used a Van Allen Summer Research Grant to show how machine learning can be to identify astronmical outliers in the Chandra Source Catalog v2.0. She developed a script in Python to query the SIMBAD astronomical database for objects near our outlier coordinates. She also considered the implications of missing data in the dataset. A follow-up study of the bias introduced by means of handling missing data has implications for future astronomical machine learning surveys.

Articles

Galaxy with X-ray grating overlaid

DeRoo Group Using AI to Identify Astronomical Outliers

In an article published in the March 2021 Iowa Magazine, the DeRoo Group was noted as one of several researchers using artificial intelligence to drive discovery. The University of Iowa team is teaching computers to categorize celestial objects on their own, then flag for further examination the phenomena that don't fit into neat categories.
Dustin Swarm Examining a Kohonen Map

Dustin Swarm Named ISGC Graduate Fellow

Dustin Swarm was named an Iowa Space Grant Consortium (ISGC) Graduate Research Fellow for 2019 - 2020