Machine Learning Tools for Open Cluster Characterization with Gaia DR2 Data

Author: Álvaro Yunta, Carlos D. (@cdalvaro)

Advisor: Guzmán-Álvarez, César A. (@cguz)

Table of Contents

• Abstract
• Data
• Code
• Acknowledgement
• References

Abstract

The characterization and understanding of Open Clusters (OCs) allow us to understand better properties and mechanisms about the Universe such as stellar formation and the regions where these events occur. They also provide information about stellar processes and the evolution of the galactic disk.

In this work, we present a novel method to characterize OCs. Our method employs a model built on Artificial Neural Networks (ANNs). More specifically, we adapted a state of the art model, the Deep Embedded Clustering (DEC) model for our purpose. The developed method aims to improve classical state of the arts techniques. We improved not only in terms of computational efficiency (with lower computational requirements), but in usability (reducing the number of hyperparameters to get a good characterization of the analyzed clusters). For our experiments, we used the Gaia DR2 database as the data source, and compared our model with the clustering technique K-Means. Our method achieves good results, becoming even better (in some of the cases) than current techniques.

Data

The data used in this project has been recovered from the Gaia Mission [1] archive. Exactly, from the DR2 dataset [2].

It is a catalogue that contains over 1.692 million registries of stars data.

In order to reduce the amount of data to be downloaded, the OPENCLUST [3] catalogue has been used to restrict the sky regions to those areas corresponding to the clusters inside the catalogue giving an extra marging to take into account stars outside clusters.

OpenClust Catalogue Distribution

Code

All code used in this project is available inside src/ directory. It is written in Python and contains downloaders, data managers, and the ML algorithms described in the thesis.

For more information see: src/README.md

Acknowledgement

This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

This publication makes use of VOSA, developed under the Spanish Virtual Observatory project supported by the Spanish MINECO through grant AyA2017-84089. VOSA has been partially updated by using funding from the European Union's Horizon 2020 Research and Innovation Programme, under Grant Agreement nº 776403 (EXOPLANETS-A)

This research has made use of the VizieR catalogue access tool, CDS, Strasbourg, France (DOI : 10.26093/cds/vizier). The original description of the VizieR service was published in 2000 (A&AS 143, 23, [4]).

References

1. G. Collaboration et al. Description of the gaia mission (spacecraft, instruments, survey and measurement principles, and operations). Gaia Collaboration et al.(2016a): Summary description of Gaia DR1, 2016. ↩️

2. C. Gaia, A. Brown, A. Vallenari, T. Prusti, J. de Bruijne, C. Babusiaux, Á. Juhász, G. Marschalkó, G. Marton, L. Molnár, et al. Gaia data release 2 summary of the contents and survey properties. Astronomy & Astrophysics, 616(1), 2018. ↩️

3. W. Dias, B. Alessi, A. Moitinho, and J. Lépine. New catalogue of optically visible open clusters and candidates. Astronomy & Astrophysics, 389(3):871–873, 2002. URL https://heasarc.gsfc.nasa.gov/W3Browse/star-catalog/openclust.html. ↩️

4. F. Ochsenbein, P. Bauer, and J. Marcout. The vizier database of astronomical catalogues. Astronomy and Astrophysics Supplement Series, 143(1):23–32, 2000. doi: 10.26093/cds/vizier. ↩️