Image Details

Choose export citation format:

A Search for a More Robust Active Galactic Nucleus Classifier Using Machine Learning

  • Authors: Alexander Messick, Vivienne Baldassare

Alexander Messick and Vivienne Baldassare 2026 The Astrophysical Journal 1009 .

  • Provider: AAS Journals

Caption: Figure 6.

The two-dimensional embedding of our data as determined by t-SNE using the “Best,” “Phot,” and “Spec” feature sets, respectively. The color of each point corresponds to active galaxies (green and yellow) and inactive galaxies (purple and blue). We also divide these populations by their masses, where “low mass” means ﹩\mathrm{log}{M}_{* }/{M}_{\odot }\lt 10﹩. We find no clear separation between the two populations on the center plot. In the first plot, we also show a crude empirical cut to the space that could be used to distinguish AGNs from non-AGNs with reasonable accuracy. Although this cut does a serviceable job separating active and inactive galaxies, t-SNE cannot be used to classify to new galaxies.

Other Images in This Article
Copyright and Terms & Conditions

Additional terms of reuse