Machine Learning Classifiers for Intermediate Redshift Emission Line Galaxies

Abstract

Classification of intermediate redshift ($z$ = 0.3--0.8) emission line galaxies as star-forming galaxies, composite galaxies, active galactic nuclei (AGN), or low-ionization nuclear emission regions (LINERs) using optical spectra alone was impossible because the lines used for standard optical diagnostic diagrams: [NII], H$\alpha$, and [SII] are redshifted out of the observed wavelength range. In this work, we address this problem using four supervised machine learning classification algorithms: $k$-nearest neighbors (KNN), support vector classifier (SVC), random forest (RF), and a multi-layer perceptron (MLP) neural network. For input features, we use properties that can be measured from optical galaxy spectra out to $z<0.8$---[OIII]/H$\beta$, [OII]/H$\beta$, [OIII] line width, and stellar velocity dispersion---and four colors ($u-g$, $g-r$, $r-i$, and $i-z$) corrected to $z=0.1$. The labels for the low redshift emission line galaxy training set are determined using standard optical diagnostic diagrams. RF has the best area under curve (AUC) score for classifying all four galaxy types, meaning highest distinguishing power. Both the AUC scores and accuracies of the other algorithms are ordered as MLP$>$SVC$>$KNN. The classification accuracies with all eight features (and the four spectroscopically-determined features only) are 93.4% (92.3%) for star-forming galaxies, 69.4% (63.7%) for composite galaxies, 71.8% (67.3%) for AGNs, and 65.7% (60.8%) for LINERs. The stacked spectrum of galaxies of the same type as determined by optical diagnostic diagrams at low redshift and RF at intermediate redshift are broadly consistent. Our publicly available code (https://github.com/zkdtc/MLC_ELGs) and trained models will be instrumental for classifying emission line galaxies in upcoming wide-field spectroscopic surveys.

arXiv comment14 pages, 14 figures, accepted for publication on ApJ

References in corpus (38)

• The Sloan Digital Sky Survey: Technical Summary
• The Host Galaxies of AGN
• The Host Galaxies and Classification of Active Galactic Nuclei
• Overview of the DESI Legacy Imaging Surveys
• K-corrections and filter transformations in the ultraviolet, optical, and near infrared
• Sloan Digital Sky Survey IV: Mapping the Milky Way, Nearby Galaxies, and the Distant Universe
• The SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Overview and Early Data
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Galaxy Zoo 1 : Data Release of Morphological Classifications for nearly 900,000 galaxies
• Extragalactic Science, Cosmology and Galactic Archaeology with the Subaru Prime Focus Spectrograph (PFS)
• The 2dF QSO Redshift Survey - I. The Optical QSO Luminosity Function
• Theoretical Evolution of Optical Strong Lines across Cosmic Time
• The MOSFIRE Deep Evolution Field (MOSDEF) Survey: Rest-Frame Optical Spectroscopy for ~1500 H-Selected Galaxies at 1.37 < z < 3.8
• Stellar population models at high spectral resolution
• The Fifteenth Data Release of the Sloan Digital Sky Surveys: First Release of MaNGA Derived Quantities, Data Visualization Tools and Stellar Library
• Semi-empirical analysis of Sloan Digital Sky Survey galaxies III. How to distinguish AGN hosts
• The MOSDEF Survey: Electron Density and Ionization Parameter at $z\sim 2.3$
• The Cosmic BPT Diagram: Confronting Theory with Observations
• A New Diagnostic of Active Galactic Nuclei: Revealing Highly-Absorbed Systems at Redshift>0.3
• 4MOST - 4-metre Multi-Object Spectroscopic Telescope
• A Survey of Galaxy Kinematics to z ~ 1 in the TKRS/GOODS-N Field. I. Rotation and Dispersion Properties
• Finding Strong Gravitational Lenses in the Kilo Degree Survey with Convolutional Neural Networks
• Prime Focus Spectrograph (PFS) for the Subaru Telescope: Overview, recent progress, and future perspectives
• Finding strong lenses in CFHTLS using convolutional neural networks
• Widespread and Hidden Active Galactic Nuclei in Star-Forming Galaxies at Redshift > 0.3
• Evolution of the Most Massive Galaxies to z=0.6: I. A New Method for Physical Parameter Estimation
• Spectral classification of emission-line galaxies from the Sloan Digital Sky Survey. I. An improved classification for high redshift galaxies
• New diagnostic methods for emission-line galaxies in deep surveys
• Finding high-redshift strong lenses in DES using convolutional neural networks
• AEGIS: Demographics of X-ray and Optically Selected AGNs
• The MOSDEF survey: AGN multi-wavelength identification, selection biases and host galaxy properties
• The Canada-France Redshift Survey XII: Nature of emission-line field galaxy population up to z = 0.3
• An automatic taxonomy of galaxy morphology using unsupervised machine learning
• LensExtractor: A Convolutional Neural Network in Search of Strong Gravitational Lenses
• Testing Diagnostics of Nuclear Activity and Star Formation in Galaxies at z>1
• The evolution of the [OII], Hβ and [OIII] emission-line luminosity functions over the last nine billions years
• Spectral classification of emission-line galaxies from the Sloan Digital Sky Survey. II. A supplementary diagnostic for AGNs using the Dn(4000) index
• Noise Estimates for Measurements of Weak Lensing from the Lyman-alpha Forest

Cited by in corpus (9)

• The Sixteenth Data Release of the Sloan Digital Sky Surveys: First Release from the APOGEE-2 Southern Survey and Full Release of eBOSS Spectra
• The final SDSS-IV/SPIDERS X-ray point source spectroscopic catalogue
• An obscured AGN population hidden in the VIPERS galaxies: identification through spectral energy distribution decomposition
• Predicting the redshift of gamma-ray loud AGNs using Supervised Machine Learning: Part 2
• Predicting the redshift of gamma-ray loud AGNs using supervised machine learning
• An Active Galactic Nucleus Recognition Model based on Deep Neural Network
• Machine learning prediction for mean motion resonance behaviour -- The planar case
• Classifying Seyfert galaxies with deep learning
• Large-step neural network for learning the symplectic evolution from partitioned data