This article explores the problem of automatically classifying astronomical objects (stars, galaxies, and quasars) based on photometric data using a neural network model. The relevance of this work stems from the rapid growth of astronomical observations, making manual classification virtually impossible, and the development of machine and deep learning methods, which offer new opportunities for automating the analysis of astronomical catalogs. The task is complicated by the high variability of photometric features and the partial overlap of characteristics between object classes. The publicly available Stellar Classification Dataset from the Kaggle platform was used as the data source. The data, obtained from the Sloan Digital Sky Survey, is a table of 100,000 objects, each described by a set of physical and observational parameters. The target feature (class label) takes one of three values. The Python programming language, as well as the Keras, TensorFlow, Scikit-learn, NumPy, Pandas, and Matplotlib libraries and frameworks, were used to build the neural network classification model. The neural network model is implemented as a multilayer perceptron, comprising three hidden layers with a ReLU activation function and a dropout mechanism to prevent overfitting. The feature space is supplemented with color indices, allowing the model to account for the shape of the object's spectrum. The constructed model achieved 97% accuracy on the validation data set. Throughout the training process, the loss values for the training and validation data sets were close. This indicates the absence of significant overfitting, as confirmed by the stabilization of both curves at approximately 0.1 by the end of the 30th training epoch. After constructing the neural network model, its effectiveness was analyzed. For this purpose, a classification error matrix was constructed, and the accuracy, precision, recall, and F1-score metrics were calculated for each class. An analysis of the error matrix and the aforementioned metrics revealed inherent difficulties in distinguishing galaxies from quasars due to overlapping color indices. This analysis also identified areas for further model improvement: expanding the feature space, applying more complex neural network architectures, adapting the model to process other datasets, and applying model interpretability methods. The practical significance of this work lies in the potential application of the developed solution for automated processing of large catalogs of astronomical observations.
ASTRONOMICAL OBJECT CLASSIFICATION, NEURAL NETWORK MODELING, MULTILAYER PERCEPTRON, PHOTOMETRIC DATA, COLOR INDICES, SDSS, MACHINE LEARNING
1. S. Wang, B. Chen, Informatics. Economics. Management, 2, 4, 134-144 (2023).
2. R.I. Yusupova, R.S. Zaripova, M.G. Nuriev, Kazan science. 2025. № 2. C. 201-203.
3. D.V. Kataseva, D.D. Loseva, Herald of Technological University, 25, 1, 77-80 (2022).
4. R.N. Minnikhanov, A.S. Katasev, D.V. Katasyova, V. Skibin, Herald of Technological University, 27 Bulletin of the, 11, 199-205 (2024).
5. K.A. Matveeva, R.N. Minnikhanov, A.S. Katasev, Herald of Technological University, 27, 1, 76-80 (2024).
6. V.A. Lokhvitsky, E.L. Yakovlev, I.V. Bushev, Intelligent Technologies in Transport, 4 (44), 89-98 (2025).
7. A.S. Khodjaev, I.I. Zinchenko, Astrophysical Bulletin, 80, 1, 145-167 (2025).
8. V.A. Golov, D.A. Petrusevich, Russian Technological Journal, 9, 3 (41), 66-77 (2021).
9. G.A. Galazutdinov, Astrophysical Bulletin, 77, 4, 579-590 (2022).
10. L. Tan, S.V. Novikova, Bulletin of the Kazan State Power Engineering University, 14, 3, 114-125 (2022).
11. K.A. Omarova, A.U. Akhmadov, O.A. Pyrnova, Problems of modern pedagogical education, 85-4, 312-317 (2024).
12. M.G. Nuriev, E.S. Belashova, K.A. Barabash, Software systems and computational methods, 1, 1 (2023).
13. K.A. Saprykin, Innovations and Investments, 11, 322-325 (2023).
14. YES. Nabiullin, V.V. Kononova, S.V. Novikova, Herald of Technological University, 24, 6, 103-107 (2021).
15. A.S. Katasev, D.V. Kataseva, Yu.N. Smirnov, L.N. Kunafina, Herald of Technological University, 27, 10, 101-105 (2024).
16. A.I. Sabirov, R.N. Minnikhanov, A.S. Katasev, R.I. Mustafin, Herald of Technological University, 27, 1, 87-91 (2024).
17. A. Bobokhonov, L. Khuramov, A. Rashidov, Problems of computer science, 3 (68), 52-71 (2025).
18. K. Lathong, K. Wisaeng, International Journal for Computational Civil and Structural Engineering, 20, 3, 69-83 (2024).
19. A.N. Grigoriev, A.I. Altukhov, D.S. Korshunov, Scientific and Technical Bulletin of Information Technologies, Mechanics and Optics, 21, 3, 311-319 (2021).
20. A.N. Tarasenkov, INASAN Science Reports, 9, 3, 89-92 (2024).
21. O.A. Pyrnova, R.S. Zaripova, Information technologies in construction, social and economic systems, 1 (15), 82-84 (2019).
22. M.P. Shleimovich, M.V. Dagaeva, A.S. Katasev, S.A. Lyasheva, M.V. Medvedev, Computer Research and Modeling, 10, 3, 369-376 (2018).
23. M.V. Dagaeva, D.V. Kataseva, A.S. Katasev, Information and Security, 21, 4, 540-545 (2018).
24. V.S. Tormozov, K.A. Vasilenko, A.L. Zolkin, Software products and systems, 2, 343-348 (2020).
25. M.I. Chernykh, S.A. Pobeda, D.A. Zhukalin, Bulletin of Voronezh State University. Series: Physics. Mathematics, 2, 42-49 (2021).
26. kaggle.com: website. – URL: https://www.kaggle.com/datasets/fedesoriano/stellarclassification-dataset-sdss17 (date of access: 25.03.2026).
27. A.R. Raju, T. Maddileti, R. Srinivas, K. Saikumar, International Journal of Integrated Engineering, 14, 7 (2022).
28. D.H. Bekov, T.A. Akhunov, O.A. Burkhonov, N.R. Alimova, Astrophysical Bulletin, 79, 1, 16-25 (2024).
29. I.Yu. Pikalov, Bulletin of Moscow State Pedagogical Univ. Series: Informatics and Informatization of Education, 2 (68), 48-61 (2024).
30. O.L. Ibryaeva, D.K. Lebedev, Automation in Industry, 8, 55-58 (2023).
31. N.A. Pastukhov, Computer-aided design in mechanical engineering, 17, 87-89 (2024).
32. B.H. Nurgaliev, A.S. Katasev, Herald of Technological University, 25, 1, 61-64 (2022).
33. E.Yu. Shmalko, Yu.A. Rumyantsev, R.R. Baynazarov, K.L. Yamshanov, Computer Science and Automation, 20, 6, 1254-1278 (2021).
34. A.S. Katasev, D.V. Kataseva, Yu.N. Smirnov, V.L. Vasiliev, Herald of Technological University, 27, 9, 115-119 (2024).
35. A.S. Katasev, D.V. Kataseva, Yu.N. Smirnov, N.I. Fattakhov, Herald of Technological University, 27, 10, 112-119 (2024).
36. D.M. Obolensky, V.I. Shevchenko, O.V. Chengar, E.N. Mashchenko, Economics. Computer Science, 48, 1, 100-115 (2021).
37. I.A. Brokarev, M.P. Farkhadov, S.V. Vaskovskii, Tomsk State University Journal of Control and Computer Science., 55, 11-17 (2021).
38. V.V. Starovoytov, Yu.I. Golub, Computer Science, 18, 1, 61-71 (2021).
39. V.A. Bulanov, O.E. Fomicheva, Devices and systems. Management, control, diagnostics, 3, 33-37 (2020).
40. L.Yu. Emaletdinova, A.S. Katasev, M.A. Nazarov, Engineering Herald of the Don, 7 (103), 71-80 (2023)
