BATEIA-1: Real and Synthetic Star Clusters

This dataset accompanies the BATEIA-1 research project (Balanced ArTificial-intelligence Method for EstImating parameters of stAr clusters), which investigates the estimation of structural parameters of open star clusters using machine learning and deep learning.

The research combines prepared stellar data from Gaia DR3, reference structural parameters from the literature, and synthetic cluster generation based on King profiles. Stellar sky maps are used for visual feature extraction, followed by regression of four structural parameters: central stellar density (f0), core radius (Rc), tidal radius (Rt), and background stellar density (fbg).

This deposit provides research data associated with the training and evaluation of the BATEIA-1 approach. Its intended uses include reproducing the associated experiments, investigating the methodology, and developing comparative approaches to open star cluster characterization.

The methodology is intended for non-embedded open clusters whose spatial distributions can be adequately described by single-component King profiles. Reference labels inherit the assumptions and limitations of the underlying profile fits and literature measurements.

Associated publication:
Mendes, D. C.; Santiago Júnior, V. A.; Angelo, M. S. (2026). BATEIA-1: An Approach for determining the structural parameters of open star clusters via machine and deep learning. Astronomy and Computing, 56, 101119.
https://doi.org/10.1016/j.ascom.2026.101119

Users should cite this dataset and the associated article when using these materials, together with the original data sources relevant to their analysis.