Automatic classification of eclipsing binary stars using deep learning methods

In the last couple of decades, tremendous progress has been achieved in developing robotic telescopes and, as a result, sky surveys (both terrestrial and space) have become source substantial amount new observational data. These data contain lot information about binary stars, hidden their light curves. With huge astronomical gathered, it is not reasonable to expect all be manually processed analyzed. Therefore, this paper, we focus on automatic classification eclipsing stars using deep learning methods. Our classifier provides tool for categorization curves into two classes: detached over-contact. We used ELISa software obtain synthetic data, which then training classifier. For evaluation purposes, collected 100 observed order evaluate number classifiers. evaluated semi-detached detached. The best-performing combines bidirectional Long Short-Term Memory (LSTM) one-dimensional convolutional neural network, 98% accuracy set. Omitting could 100% classification.