Keywords
structure and operation
studying
optical spectra
recognition
physical laboratory work
How to Cite
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.
References
T. S. Francis, H. Y. Edward, and G. Y. Ann, The art of learning: neural networks and education. Boca Raton, FL, USA: CRC Press, 2019.
В. Биков, О. Спірін, та О. Пінчук, “Cучасні завдання цифрової трансформації освіти”, Вісник Кафедри ЮНЕСКО “Неперервна професійна освіта ХХІ століття”, вип. 1, с. 27–36, Трав. 2020. doi: 10.35387/ucj.1(1).2020.27-36.
A. Gupta, V. Aggarwal, V. Sharma and M. Naved, “Framework to Integrate Education 4.0 to Enhance the E-Learning Model for Industry 4.0 and Society 5.0” in The Role of Sustainability and Artificial Intelligence in Education Improvement, J. Rosak-Szyrocka, J. Żywiołek, A. Nayyar and M. Naved, Eds., Boca Raton, FL, USA: Chapman and Hall/CRC, 2024, pp. 151–167.
D. Passos, and P. Mishra, “A tutorial on automatic hyperparameter tuning of deep spectral modelling for regression and classification tasks”, Chemometrics and Intelligent Laboratory Systems, vol. 223, Apr. 15, 2022, Art. no. 104520, doi: 10.1016/j.chemolab.2022.104520.
L. Hannemose Rieger, M. Wilson, T. Vegge, and E. Flores, “Understanding the patterns that neural networks learn from chemical spectra”, Digital Discovery, no. 2, pp. 1957–1968, Nov. 2023, doi: 10.1039/D3DD00203A.
J. Liu, M. Osadchy, L. Ashton, M. Foster, C. J. Solomone, and S. J. Gibson, “Deep convolutional neural networks for Raman spectrum recognition: a unified solution”, Analyst, vol. 142, pp. 4067–4074, Sep. 2017, doi: 10.1039/C7AN01371J.
P. Jahoda, I. Drozdovskiy, S. J. Payler, L. Turchi, L. Bessone, and F. Sauro, “Machine learning for recognizing minerals from multispectral data”, Analyst, vol. 146, pp. 184–195, Oct. 2021, doi: 10.1039/D0AN01483D.
X. Fan, W. Ming, H. Zeng, Z. Zhang, and H. Lu, “Deep learning-based component identification for the Raman spectra of mixtures”, Analyst, vol. 144, pp. 1789–1798, Jan. 2019, doi: 10.1039/C8AN02212G.
R. Zhang et al., “Transfer-learning-based Raman spectra identification”, J. Raman Spectrosc., vol. 51, pp. 176–186, Nov. 2020, doi: 10.1002/jrs.5750.
W. Zhou, Y. Tang, Z. Qian, J. Wang, and H. Guo, “Deeply-recursive convolutional neural network for Raman spectra identification”, RSC Advances, Vol. 12, pp. 5053–5061, Feb. 2022, doi 10.1039/D1RA08804A.
J. Acquarelli, T. van Laarhoven, J. Gerretzen, T. N. Tran, Lutgarde M. C. Buydens, and E. Marchiori, “Convolutional neural networks for vibrational spectroscopic data analysis”, Analytica Chimica Acta, vol. 954, pp. 22–31, Feb. 2017, doi: https://doi.org/10.1016/j.aca.2016.12.010.
D. J. Brown, K. D. Shepherd, M. G. Walsh, M. D. Mays, and T. G. Reinschc, “Global soil characterization with VNIR diffuse reflectance spectroscopy”, Geoderma, vol. 132, no. 3–4, pp. 273–290, Jun. 2006, doi: 10.1016/j.geoderma.2005.04.025.
D. Prochazka, P. Porizka, J. Hruska, K. Novotny, A. Hrdlicka, and J. Kaiser, “Machine learning in laser-induced breakdown spectroscopy as a novel approach towards experimental parameter optimization”, J. Analytical Atomic Spectrometry, vol. 37, pp. 603–612, Feb. 2022, doi: 10.1039/D1JA00389E.
Y. Fan et al., “Deep learning-based method for automatic resolution of gas chromatography-mass spectrometry data from complex samples”, Journal of Chromatography A, vol. 1690, Feb. 8, 2023, Art. no. 463768, doi: 10.1016/j.chroma.2022.463768.
G. Huab, and M. Qiu, “Machine learning-assisted structure annotation of natural products based on MS and NMR data”, Nat. Prod. Rep., vol. 40, pp. 1735–1753, Jul. 2023, doi: 10.1039/D3NP00025G.
Z. Chen, Y. Xie, Y. Wu, Y. Lin, S. Tomiya, and J. Lin, “An interpretable and transferrable vision transformer model for rapid materials spectra classification”, Digital Discovery, no. 3, pp. 369–380, 2024, doi: 10.1039/D3DD00198A.
R. Zhao, D. Wu, J. Wen, Q. Zhang, G. Zhang, and J. Li, “Robustness and accuracy improvement of data processing with 2D neural networks for transient absorption dynamics”, Phys. Chem. Chem. Phys., vol. 23, pp. 16998–17008, Jul. 2021, doi: 10.1039/D1CP02521J.
С. О. Субботін, Нейронні мережі: теорія та практика: навчальний посібник. Житомир, Україна: Вид. О. О. Євенок, 2020.
S. Haykin, Neural Networks. A Comprehensive Foundations. Avrora, N.Y., USA: McMillan College Publ. Co., 1994.
T. Rashid, Make Your Own Neural Network. Charleston, SC, USA: CreateSpace Independent Publishing Platform, 2016.
С. П. Величко, І. С. Величко, С. Г. Ковальов, та В. В. Миколайко, “Створення сучасного комплекту для вивчення оптичного випромінювання у практикумі з фізики в закладах вищої освіти”, у MODERNÍ ASPEKTY VĚDY: Svazek XXVIII mezinárodní kolektivní monografie, Есеніце, Чеська Республіка: Mezinárodní Ekonomický Institut s.r.o., 2023, С. 170–271.
С. Величко, І. Величко, та С. Ковальов, «Створення сучасного спектрального приладу для навчальних цілей», Актуальні питання у сучасній науці, № 1(7), С. 338–351, Jan. 2023. doi: 10.52058/2786-6300-2023-1(7)-338-351.
REFERENCES (TRANSLATED AND TRANSLITERATED)
T. S. Francis, H. Y. Edward and G. Y. Ann, The art of learning: neural networks and education. Boca Raton, FL, USA: CRC Press, 2019. (in English).
V. Bykov, O. Spirin, and O. Pinchuk, «Modern tasks of digital transformation of education», UNESCO Chair Journal Lifelong Professional Education in the XXI Century, no. 1, pp. 27-36, May 2020. doi:10.35387/ucj.1(1).2020.27-36. (in Ukrainian).
A. Gupta, V. Aggarwal, V. Sharma and M. Naved, «Framework to Integrate Education 4.0 to Enhance the E-Learning Model for Industry 4.0 and Society 5.0» in The Role of Sustainability and Artificial Intelligence in Education Improvement, J. Rosak-Szyrocka, J. Żywiołek, A. Nayyar and M. Naved, Eds., Boca Raton, FL, USA: Chapman and Hall/CRC, 2024, pp. 151–167. (in English).
D. Passos, and P. Mishra, “A tutorial on automatic hyperparameter tuning of deep spectral modelling for regression and classification tasks”, Chemometrics and Intelligent Laboratory Systems, vol. 223, Apr. 15, 2022, Art. no. 104520, doi: 10.1016/j.chemolab.2022.104520. (in English).
L. Hannemose Rieger, M. Wilson, T. Vegge, and E. Flores, “Understanding the patterns that neural networks learn from chemical spectra”, Digital Discovery, no. 2, pp. 1957–1968, Nov. 2023, doi: 10.1039/D3DD00203A. (in English).
J. Liu, M. Osadchy, L. Ashton, M. Foster, C. J. Solomone, and S. J. Gibson, “Deep convolutional neural networks for Raman spectrum recognition: a unified solution”, Analyst, vol. 142, pp. 4067–4074, Sep. 2017, doi: 10.1039/C7AN01371J. (in English).
P. Jahoda, I. Drozdovskiy, S. J. Payler, L. Turchi, L. Bessone, and F. Sauro, “Machine learning for recognizing minerals from multispectral data”, Analyst, vol. 146, pp. 184–195, Oct. 2021, doi: 10.1039/D0AN01483D. (in English).
X. Fan, W. Ming, H. Zeng, Z. Zhang, and H. Lu, “Deep learning-based component identification for the Raman spectra of mixtures”, Analyst, vol. 144, pp. 1789–1798, Jan. 2019, doi: 10.1039/C8AN02212G.
R. Zhang et al., “Transfer-learning-based Raman spectra identification”, J. Raman Spectrosc., vol. 51, pp. 176–186, Nov. 2020, doi: 10.1002/jrs.5750. (in English).
W. Zhou, Y. Tang, Z. Qian, J. Wang, and H. Guo, “Deeply-recursive convolutional neural network for Raman spectra identification”, RSC Advances, Vol. 12, pp. 5053–5061, Feb. 2022, doi 10.1039/D1RA08804A. (in English).
J. Acquarelli, T. van Laarhoven, J. Gerretzen, T. N. Tran, Lutgarde M. C. Buydens, and E. Marchiori, “Convolutional neural networks for vibrational spectroscopic data analysis”, Analytica Chimica Acta, vol. 954, pp. 22–31, Feb. 2017, doi: https://doi.org/10.1016/j.aca.2016.12.010. (in English).
D. J. Brown, K. D. Shepherd, M. G. Walsh, M. D. Mays, and T. G. Reinschc, “Global soil characterization with VNIR diffuse reflectance spectroscopy”, Geoderma, vol. 132, no. 3–4, pp. 273–290, Jun. 2006, doi: 10.1016/j.geoderma.2005.04.025. (in English).
D. Prochazka, P. Porizka, J. Hruska, K. Novotny, A. Hrdlicka, and J. Kaiser, “Machine learning in laser-induced breakdown spectroscopy as a novel approach towards experimental parameter optimization”, J. Analytical Atomic Spectrometry, vol. 37, pp. 603–612, Feb. 2022, doi: 10.1039/D1JA00389E. (in English).
Y. Fan et al., “Deep learning-based method for automatic resolution of gas chromatography-mass spectrometry data from complex samples”, Journal of Chromatography A, vol. 1690, Feb. 8, 2023, Art. no. 463768, doi: 10.1016/j.chroma.2022.463768. (in English).
G. Huab, and M. Qiu, “Machine learning-assisted structure annotation of natural products based on MS and NMR data”, Nat. Prod. Rep., vol. 40, pp. 1735–1753, Jul. 2023, doi: 10.1039/D3NP00025G. (in English).
Z. Chen, Y. Xie, Y. Wu, Y. Lin, S. Tomiya, and J. Lin, “An interpretable and transferrable vision transformer model for rapid materials spectra classification”, Digital Discovery, no. 3, pp. 369–380, 2024, doi: 10.1039/D3DD00198A. (in English).
R. Zhao, D. Wu, J. Wen, Q. Zhang, G. Zhang, and J. Li, “Robustness and accuracy improvement of data processing with 2D neural networks for transient absorption dynamics”, Phys. Chem. Chem. Phys., vol. 23, pp. 16998–17008, Jul. 2021, doi: 10.1039/D1CP02521J. (in English).
S. O. Subbotin, Neural networks: theory and practice: a study guide. Zhytomyr, Ukraine: O. O. Evenok Publishing House, 2020. (in Ukrainian).
S. Haykin, Neural Networks. A Comprehensive Foundations. Avrora, N.Y., USA: McMillan College Publ. Co., 1994. (in English).
T. Rashid, Make Your Own Neural Network. Charleston, SC, USA: CreateSpace Independent Publishing Platform, 2016. (in English).
S. P. Velychko, I. S. Velychko., S. G. Kovalov, and V. V. Mykolayko, «Creation of a modern set for the study of optical radiation in a physics workshop in institutions of higher education», in Modern aspects of science: Volume XXVIII of the international collective monograph, Jesenice, Czech Republic: International Economic Institute s.r.o., 2023, pp. 170–271. (in Ukrainian).
S. Velichko, I. Velichko, and S. Kovalov, «Creating a modern spectral device for educational purposes», Current issues in modern science, no. 1(7), pp. 338-351, Jan. 2023. doi:10.52058/2786-6300-2023-1(7)-338–351. (in Ukrainian).
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright (c) 2024 Stepan Velychko, Ihor Velychko, Serhii Kovalov, Yurii Kovalov