Photometric classification of QSOs from ALHAMBRA survey using random forest

Abstract

Context: Given the current big data era in Astronomy, machine learning based methods have being applied over the last years to identify or classify objects like quasars, galaxies and stars from full sky photometric surveys. Aims: Here we systematically evaluate the performance of Random Forests (RF) in classifying quasars using either magnitudes or colours, both from broad and narrow-band filters, as features. Methods: The working data consists of photometry from the ALHAMBRA Gold Catalogue that we cross-matched with the Sloan Digital Sky Survey (SDSS) and with the Million Quasars Catalogue (Milliquas) for objects labelled as quasars, galaxies or stars. A RF classifier is trained and tested to evaluate the effect on final accuracy and precision of varying the free parameters and the effect of using narrow or broad-band magnitudes or colours. Results: Best performances of the classifier yielded global accuracy and quasar precision around 0.9. Varying model free parameters (within reasonable ranges of values) has no significant effects on the final classification. Using colours instead of magnitudes as features results in better performances of the classifier, especially using colours from the ALHAMBRA Survey. Colours that contribute the most to the classification are those containing the near-infrared $JHK$ bands.

7 pages including 6 figures and 3 tables. Accepted for publication in Astronomy and Astrophysics

References in corpus (8)

• The ALHAMBRA Project: A large area multi medium-band optical and NIR photometric survey
• Robust Machine Learning Applied to Astronomical Datasets I: Star-Galaxy Classification of the SDSS DR3 Using Decision Trees
• Machine-learning identification of galaxies in the WISExSuperCOSMOS all-sky catalogue
• On the discovery of stars, quasars, and galaxies in the Southern Hemisphere with S-PLUS DR2
• Random Forests as a viable method to select and discover high redshift quasars
• The Extremely Luminous Quasar Survey in the Pan-STARRS 1 Footprint (PS-ELQS)
• Identification of BASS DR3 Sources as Stars, Galaxies and Quasars by XGBoost
• The Probabilistic Random Forest applied to the selection of quasar candidates in the QUBRICS Survey