Photometric Selection of Type 1 Quasars in the XMM-LSS Field with Machine Learning and the Disk–Corona Connection

We present photometric selection of type 1 quasars in the ≈5.3 deg2 XMM-Large Scale Structure survey field with machine learning. We constructed our training and blind-test samples using spectroscopically identified Sloan Digital Sky Survey quasars, galaxies, and stars. We utilized the XGBoost machine learning method to select a total of 1591 quasars. We assessed the classification performance based on the blind-test sample, and the outcome was favorable, demonstrating high reliability (≈99.9%) and good completeness (≈87.5%). We used XGBoost to estimate photometric redshifts of our selected quasars. The estimated photometric redshifts span a range from 0.41 to 3.75. The outlier fraction of these photometric redshift estimates is ≈17%, and the normalized median absolute deviation (σNMAD) is ≈0.07. To study the quasar disk–corona connection, we constructed a subsample of 1016 quasars with Hyper Suprime-Cam i < 22.5 after excluding radio-loud and potentially X-ray-absorbed quasars. The relation between the optical-to-X-ray power-law slope parameter (αOX) and the 2500 Å monochromatic luminosity (L2500Å) for this subsample is with a dispersion of 0.159. We found this correlation in good agreement with the correlations in previous studies. We explored several factors, which may bias the αOX–L2500Å relation, and found that their effects are not significant. We discussed possible evolution of the αOX–L2500Å relation with respect to L2500Å or redshift.