Learning from the machine: interpreting machine learning algorithms for point- and extended- source classification
arXiv:1712.03970 · doi:10.1093/mnras/sty2575
Abstract
We investigate star-galaxy classification for astronomical surveys in the context of four methods enabling the interpretation of black-box machine learning systems. The first is outputting and exploring the decision boundaries as given by decision tree based methods, which enables the visualization of the classification categories. Secondly, we investigate how the Mutual Information based Transductive Feature Selection (MINT) algorithm can be used to perform feature pre-selection. If one would like to provide only a small number of input features to a machine learning classification algorithm, feature pre-selection provides a method to determine which of the many possible input properties should be selected. Third is the use of the tree-interpreter package to enable popular decision tree based ensemble methods to be opened, visualized, and understood. This is done by additional analysis of the tree based model, determining not only which features are important to the model, but how important a feature is for a particular classification given its value. Lastly, we use decision boundaries from the model to revise an already existing method of classification, essentially asking the tree based method where decision boundaries are best placed and defining a new classification method. We showcase these techniques by applying them to the problem of star-galaxy separation using data from the Sloan Digital Sky Survey (hereafter SDSS). We use the output of MINT and the ensemble methods to demonstrate how more complex decision boundaries improve star-galaxy classification accuracy over the standard SDSS frames approach (reducing misclassifications by up to ). We then show how tree-interpreter can be used to explore how relevant each photometric feature is when making a classification on an object by object basis.
arXiv comment12 pages, 8 figures, 8 tables
References in corpus (20)
- Scikit-learn: Machine Learning in Python
- The Sloan Digital Sky Survey: Technical Summary
- The Eleventh and Twelfth Data Releases of the Sloan Digital Sky Survey: Final Data from SDSS-III
- The 2.5 m Telescope of the Sloan Digital Sky Survey
- Parametric Recovery of Line-of-Sight Velocity Distributions from Absorption-Line Spectra of Galaxies via Penalized Likelihood
- SDSS-III: Massive Spectroscopic Surveys of the Distant Universe, the Milky Way Galaxy, and Extra-Solar Planetary Systems
- The Baryon Oscillation Spectroscopic Survey of SDSS-III
- The Sloan Digital Sky Survey Photometric Camera
- The u'g'r'i'z' Standard Star Network
- The Second Data Release of the Sloan Digital Sky Survey
- The Multi-Object, Fiber-Fed Spectrographs for SDSS and the Baryon Oscillation Spectroscopic Survey
- The SAURON project - V. Integral-field emission-line kinematics of 48 elliptical and lenticular galaxies
- Spectral Classification and Redshift Measurement for the SDSS-III Baryon Oscillation Spectroscopic Survey
- Photometric Response Functions of the SDSS Imager
- Star/galaxy separation at faint magnitudes: Application to a simulated Dark Energy Survey
- Feature importance for machine learning redshifts applied to SDSS galaxies
- A Simple Likelihood Method for Quasar Target Selection
- The SDSS SkyServer: Public Access to the Sloan Digital Sky Server Data
- The SDSS SkyServer, Public Access to the Sloan Digital Sky Server Data
- MINT: Mutual Information based Transductive Feature Selection for Genetic Trait Prediction
Cited by in corpus (9)
- Identifying galaxies, quasars, and stars with machine learning: A new catalogue of classifications for 111 million SDSS sources without spectra
- Lyman Alpha Reference Sample: X. Predicting Lyman alpha output from starforming galaxies using multivariate regression
- Data mining techniques on astronomical spectra data. II : Classification Analysis
- Machine Learning for Observational Cosmology
- The PAU Survey: star-galaxy classification with multi narrow-band data
- Self-consistent redshift estimation using correlation functions without a spectroscopic reference sample
- A search for dark matter among Fermi-LAT unidentified sources with systematic features in Machine Learning
- Evolutionary model discovery of causal factors behind the socio-agricultural behavior of the ancestral Pueblo
- FNet II: Spectral Classification of Quasars, Galaxies, Stars, and broad absorption line (BAL) Quasars