The use of neural networks for the recognition of optical spectra in the educational practice on the fundamentals of spectral analysis

- Величко, Степан (orcid.org/0000-0002-1692-9741), Величко, Ігор (orcid.org/0000-0003-0785-4538), Ковальов, Сергій (orcid.org/0000-0002-3922-8697) and Ковальов, Юрій (orcid.org/0000-0002-1729-2033) (2024) The use of neural networks for the recognition of optical spectra in the educational practice on the fundamentals of spectral analysis Information Technologies and Learning Tools, 5 (103). pp. 174-187. ISSN 2076-8184

Text Степан Величко.pdf - Published Version Download (1MB)

Abstract

The article analyzes the idea of applying artificial intelligence in the form of an artificial neural network model for automatic recognition of optical spectra registered by traditional spectral equipment. The material presented pays special attention to the logic of constructing the structure of neural networks aimed at extracting featured for differentiating optical spectra of various chemical elements. The discussed material involves studying the effectiveness of the proposed neural network structures based on implementing their training using the gradient descent method. The authors consider the peculiarities of applying neural networks for recognizing optical spectra both in methodological terms and in the creation of a corresponding department or center in higher education institutions of Ukraine, where research work and laboratory practice in spectral analysis and spectroscopy are conducted. The workshop created this way involves familiarizing students with the basics of spectroscopy, carrying out a series of important works in this field of human activity, and introducing them to such a powerful modern research tool as artificial intelligence in the form of an artificial neural network with programming elements based on Python in the course of educational physical experiments. The operation of the laboratory practice based on the use of an artificial neural network model is analyzed, providing students with elementary understanding of the structure and functioning of this model and the possibilities of implementing it in the study of spectroscopic regularities in the optical range, as well as the possibility of establishing a department or center for studying the basics of spectral analysis spectroscopy within the higher education institution in Ukraine. A laboratory work has been proposed, which involves the study of methods for automatic classification of experimental data using the example of recognizing the optical spectra of helium and neon.

Item Type:	Article
Keywords:	artificial neural network; structure and operation; studying; optical spectra; recognition; physical laboratory work.
Subjects:	Science and knowledge. Organization. Computer science. Information. Documentation. Librarianship. Institutions. Publications > 00 Prolegomena. Fundamentals of knowledge and culture. Propaedeutics > 004 Computer science and technology. Computing. Data processing > 004.9 ІКТ ( Application-oriented computer-based techniques ) Science and knowledge. Organization. Computer science. Information. Documentation. Librarianship. Institutions. Publications > 3 Social Sciences > 37 Education > 378 Higher education. Universities. Academic study Science and knowledge. Organization. Computer science. Information. Documentation. Librarianship. Institutions. Publications > 5 Мathematics. natural sciences > 53 Physics
Divisions:	Institute for Digitalisation of Education > Generic resouse
Depositing User:	Алла 1 Алла Почтарьова
Date Deposited:	17 Jan 2025 12:13
Last Modified:	17 Jan 2025 12:13
URI:	https://lib.iitta.gov.ua/id/eprint/744254

Downloads

Actions (login required)

View Item