Yu-Yen Chang (NCHU/ASIAA), Bau-Ching Hsieh (ASIAA), Wei-Hao Wang (ASIAA), Yen-Ting Lin (ASIAA), Chen-Fatt Lim (NTU/ASIAA), Yoshiki Toba (Kyoto U/ASIAA/Ehime U) et al.
We use machine learning techniques to investigate active galaxies, including X-ray selected AGNs (XAGNs), infrared selected AGNs (IRAGNs), and radio selected AGNs (RAGNs). Using known physical parameters in the Cosmic Evolution Survey (COSMOS) field, we are able to have well-established training samples in the ultra-deep regions of Hyper Suprime-Cam (HSC) survey. We use Python packages (XGBoost and Keras) to identify AGNs and show their performance (e.g., accuracy, precision, recall, F1-score, and AUROC). Our results indicate that the performance is high for bright XAGN and IRAGN host galaxies. HSC (optical) information with Wide-field Infrared Survey Explorer (WISE) band-1 and WISE band-2 (near-infrared) information perform well to identify AGN hosts. For both type-1 (broad-band) XAGNs and type-1 (unobscured) IRAGNs, the performance is very good by using optical to infrared information. These results can apply to the five-band data from the wide regions of the HSC survey, and future all-sky surveys.