Machine Learning Classification of SDSS Transient Survey Images

TL;DR

This paper demonstrates that machine learning algorithms can effectively classify transient astronomical images from SDSS, achieving performance comparable to humans, which is crucial for future large-scale surveys like LSST.

Contribution

The study introduces PCA-based machine learning methods that match human classification performance on SDSS transient images, paving the way for automated analysis in upcoming surveys.

Findings

Achieved 96% recall in classifying real transients

Misclassified only 18% of artefacts as real objects

Random forests outperformed other algorithms

Abstract

We show that multiple machine learning algorithms can match human performance in classifying transient imaging data from the Sloan Digital Sky Survey (SDSS) supernova survey into real objects and artefacts. This is a first step in any transient science pipeline and is currently still done by humans, but future surveys such as the Large Synoptic Survey Telescope (LSST) will necessitate fully machine-enabled solutions. Using features trained from eigenimage analysis (principal component analysis, PCA) of single-epoch g, r and i-difference images, we can reach a completeness (recall) of 96 per cent, while only incorrectly classifying at most 18 per cent of artefacts as real objects, corresponding to a precision (purity) of 84 per cent. In general, random forests performed best, followed by the k-nearest neighbour and the SkyNet artificial neural net algorithms, compared to other methods such as na\"ive Bayes and kernel support vector machine. Our results show that PCA-based machine learning can match human success levels and can naturally be extended by including multiple epochs of data, transient colours and host galaxy information which should allow for significant further improvements, especially at low signal-to-noise.