A model by ANN method for contact binary stars

W Ursae Majoris (W UMa)-type binary systems are important astrophysical tools for studying star formation, stellar structure, and evolution in astrophysics. In contact binary systems, determining the mass of the primary star without using an expensive spectroscopic approach has always been a challenge. In this project, we attempted for the first time to present a method for trying to solve this problem utilizing creative and AI methods. We have obtained a model between the orbital period, the mass of the primary component, and temperature for these types of binary systems using the artificial neural network (ANN) method. We also had a comparison between our ANN model results and previous spectroscopy-based investigations for the primary component masses; the results were very close. The results of this study have been published in the prestigious Monthly Notices of the Royal Astronomical Society (MNRAS) journal. This paper is available in this link:

Machine Learning Classification for Delta-Scuti Stars

A project in the field of Delta-Scuti variable stars was carried out in collaboration with researchers from four countries. In August 2021, the results of the study were published as a peer-reviewed publication in the Publications of the Astronomical Society of the Pacific (PASP) journal. The P-L relationship was investigated in this project, and measurement accuracy was improved. We applied machine learning classification and fitted lines to the fundamental and overtone modes areas to recognize and classify data. As a result, we provide a new P-L relation for Scuti variables' fundamental and overtone modes. This paper is available in this link: The department's team came up with the idea of employing a learning machine, carried out the entire procedure, wrote the content for this section, and presented new relationships.