The miniJPAS survey quasar selection: III. Classification with artificial neural networks and hybridisation

Abstract

This paper is part of large effort within the J-PAS collaboration that aims to classify point-like sources in miniJPAS, which were observed in 60 optical bands over ~1 deg2 in the AEGIS field. We developed two algorithms based on artificial neural networks (ANN) to classify objects into four categories: stars, galaxies, quasars at low redshift (z < 2.1), and quasars at high redshift (z = 2.1). As inputs, we used miniJPAS fluxes for one of the classifiers (ANN1) and colours for the other (ANN2). The ANNs were trained and tested using mock data in the first place. We studied the effect of augmenting the training set by creating hybrid objects, which combines fluxes from stars, galaxies, and quasars. Nevertheless, the augmentation processing did not improve the score of the ANN. We also evaluated the performance of the classifiers in a small subset of the SDSS DR12Q superset observed by miniJPAS. In the mock test set, the f1-score for quasars at high redshift with the ANN1 (ANN2) are 0.99 (0.99), 0.93 (0.92), and 0.63 (0.57) for 17 < r = 20, 20 < r = 22.5, and 22.5 < r = 23.6, respectively, where r is the J-PAS rSDSS band. In the case of low-redshift quasars, galaxies, and stars, we reached 0.97 (0.97), 0.82 (0.79), and 0.61 (0.58); 0.94 (0.94), 0.90 (0.89), and 0.81 (0.80); and 1.0 (1.0), 0.96 (0.94), and 0.70 (0.52) in the same r bins. In the SDSS DR12Q superset miniJPAS sample, the weighted f1-score reaches 0.87 (0.88) for objects that are mostly within 20 < r = 22.5. We find that the most common confusion occurs between quasars at low redshift and galaxies in mocks and miniJPAS data. We discuss the origin of this confusion, and we show examples in which these objects present features that are shared by both classes. Finally, we estimate the number of point-like sources that are quasars, galaxies, and stars in miniJPAS.