UDC 004.056
The article discusses the software implementation of a mathematical classification model for identifying human attention states. The main focus is on the method of labeling the dynamics of attention changes as a time series, which allows for real-time analysis of attention using third-party software. The purpose of the study is to develop an artificial intelligence classification model for automatically determining various states of attention based on mathematical modeling of time series. The methodology is based on the application of classical machine learning methods and statistical time series analysis. The paper presents a classification algorithm, data preprocessing methods, and approaches to labeling time intervals. In the article, the authors consider in detail the developed software, its functionality and the principle of operation for reading and systematizing the database. At the end of the article, the authors discuss the experimental results, analyze the error level of the classification model, and compare the results with the results of past experience. The results of the study demonstrate the possibility of creating accurate and reliable attention monitoring systems, which opens up new perspectives in the field of cognitive research and practical psychology.
mathematical modeling, binary classification, time series, labeling, software implementation, artificial intelligence, neural interface, NCI
1. Nesterova, E. N. The influence of physical activity on the cognitive abilities of students: a study of the relationship between physical exercise and improvement of memory, concentration and general academic performance / E. N. Nesterova, M. I. Kuznetsova, I. V. Shelegin // Bulletin of Science. – 2024. – Vol. 5, No. 12-1(81). – pp. 1180-1188. – EDN QLVKTO.
2. Gilyazetdinov, A. O. Neural networks / A. O. Gilyazetdinov, N. A. Petrov // PROBLEMS and TRENDS OF SCIENTIFIC TRANSFORMATIONS in THE CONTEXT OF SOCIETY TRANSFORMATION: a collection of articles based on the results of the International Scientific and Practical Conference, Magnitogorsk, April 04, 2021. – Sterli-tamak: Limited Liability Company "Agency for International Studies", 2021. – pp. 97-99. – EDN MIOXMU.
3. A brief overview of the machine learning algorithm for the support vector machine (SVM) method / [Electronic resource] // Habr : [website]. — URL:
4. Intelligent analysis of neurointerface data by reference / E. A. Mendeleev, K. R. Petrova, D. Ya. Bortseva, R. V. Petrov // Virtual reality technologies: production, science, education (TVR-2023): Collection of articles of the All-Russian Scientific Conference, Veliky Novgorod, April 25-27, 2023. – Veliky Novgorod: Yaroslav the Wise Novgorod State University, 2024. – pp. 72-81. – EDN DOSGEZ.
5. Volkova E.V., Khlebnikova E.V. Software implementation of the classification mathematical model of identification of attention states by marking time series of dynamics of attention changes. // Neurocomputers: development, application. - 2023. — No. 12. — pp. 10-18.
6. Kulikov A.A., Ivanov V.N. Development of a classification model for identifying changes in attention levels by the method of temporal marking. // Artificial intelligence and decision-making. — 2022. — No. 3. — pp. 45-53.
7. Zakharov A.I., Kolesnikov P.S. An algorithm for automatic marking of time series to detect cognitive fatigue and changes in attention. // Bio-medical technologies and radio electronics. - 2021. — No. 5. — pp. 23-31.
8. Seliverstov A.V., Karpenko A.N. Application of artificial neural networks for recognition and labeling of attention states by time series of brain activity. // Intelligent systems and technologies. - 2023. — No. 2. — pp. 34-43.
9. Vasilyeva A.S., Tikhomirov D.A. Identification of attention states based on marker time series and deep learning. // Computer Science and Computational Engineering. — 2022. — No. 6. — pp. 55-64.
10. Yakovlev V.A., Gorelov A.N. Automated classification of attention states using labeled time series. // Telecommunications and informatization of education. — 2021. — No. 1. — pp. 20-29.
11. Smith A.B., Johnson S.P. Software implementation of a classification model for identifying states of attention by marking time series. // Expert systems with applications. — 2023. — Volume 123, pp. 15-23.
12. Lee S.Y., Kim H.J. Development of a classifier based on time series markup to identify changes in attention levels. // Neurocomputers. - 2022. — Volume 493, pp. 234-242.
13. Williams D.G., Davis J.T. An algorithm for automatic processing of time series data to detect cognitive fatigue and attention shifts. // IEEE Transactions on Neural Networks and Learning Systems. — 2021. — Volume 32, No. 5, pp. 1876-1885.
14. Kumar R., Singh A. Application of artificial neural networks for recognition and labeling of attention states by time series of brain activity. // Computational intelligence and neuroscience. — 2023. — Volume 2023, Article 4367854, 12 pages.
15. Petersen M., Olsson H. Identification of attention states using time series based on markers and deep learning algorithms. // Image recognition. — 2022. — Volume 125, p. 108456.
16. Nguyen T.K., Tran L.V. Automated classification of attention states using the time series labeling method. // Knowledge-based systems. — 2021. — Volume 213, p. 106638.
