An improved quasar detection method in EROS-2 and MACHO LMC datasets

TL;DR

This paper introduces a boosted Random Forest classifier utilizing auto regressive features for more accurate quasar detection in large astronomical datasets, achieving around 90% precision and 86% recall.

Contribution

The study presents a novel classification approach that significantly improves quasar detection accuracy in EROS-2 and MACHO datasets using auto regressive parameters.

Findings

Achieved approximately 90% precision and 86% recall in training datasets.

Identified 1160 and 2551 quasar candidates in EROS-2 and MACHO datasets.

Validated candidates with a 74% and 40% match rate to known quasars.

Abstract

We present a new classification method for quasar identification in the EROS-2 and MACHO datasets based on a boosted version of Random Forest classifier. We use a set of variability features including parameters of a continuous auto regressive model. We prove that continuous auto regressive parameters are very important discriminators in the classification process. We create two training sets (one for EROS-2 and one for MACHO datasets) using known quasars found in the LMC. Our model's accuracy in both EROS-2 and MACHO training sets is about 90% precision and 86% recall, improving the state of the art models accuracy in quasar detection. We apply the model on the complete, including 28 million objects, EROS-2 and MACHO LMC datasets, finding 1160 and 2551 candidates respectively. To further validate our list of candidates, we crossmatched our list with a previous 663 known strong candidates, getting 74% of matches for MACHO and 40% in EROS-2. The main difference on matching level is because EROS-2 is a slightly shallower survey which translates to significantly lower signal-to-noise ratio lightcurves.