This paper demonstrates a novel and efficient unsupervised clustering method with the combination of a self-organizing map (SOM) and a convolutional autoencoder. The rapidly increasing volume of radio-astronomical data has increased demand for machine-learning methods as solutions to classification and outlier detection. Major astronomical discoveries are unplanned and found in the unexpected, making unsupervised machine learning highly desirable by operating without assumptions and labeled training data. Our approach shows SOM training time is drastically reduced and high-level features can be clustered by training on auto-encoded feature vectors instead of raw images. Our results demonstrate this method is capable of accurately separating outliers on a SOM with neighborhood similarity and K-means clustering of radio-astronomical features. We present this method as a powerful new approach to data exploration by providing a detailed understanding of the morphology and relationships of Radio Galaxy Zoo (RGZ) data set image features which can be applied to new radio survey data.
|Original language||English (US)|
|Journal||Publications of the Astronomical Society of the Pacific|
|State||Published - Oct 2019|
Bibliographical noteFunding Information:
The authors acknowledge the Radio Galaxy Zoo Project builders and volunteers listed in full athttp://rgzauthors. galaxyzoo.org for their contribution to RGZ data set and labels used in this paper. We also acknowledge the National Radio Astronomy Observatory (NRAO) and the Karl G. Jansky Very Large Array (VLA) as the source of this radio data. Partial support for L.R is provided by the U.S National Science Foundation grant AST17-14205 to the University of Minnesota. H.A benefited from grant DAIP #066/2018 of Universidad de Guanajuato.
© 2019. The Astronomical Society of the Pacific.
- Astronomical databases: miscellaneous
- Methods: data analysis
- Radio continuum: galaxies