Simulation of sports movement training based on machine learning and brain-computer interface

Issue title: Fuzzy model for human autonomous computing in extreme surveillance and it’s applications

Guest editors: Varatharajan Ramachandran

Article type: Research Article

Authors: Wang, Linuo^{; *}

Affiliations: Xi’an University of Finance and Economics, Shaanxi, China

Correspondence: [*] Corresponding author. Linuo Wang, Xi’an University of Finance and Economics, Shaanxi, China. E-mail: [email protected].

Abstract: Injuries and hidden dangers in training have a greater impact on athletes ’careers. In particular, the brain function that controls the motor function area has a greater impact on the athlete ’s competitive ability. Based on this, it is necessary to adopt scientific methods to recognize brain functions. In this paper, we study the structure of motor brain-computer and improve it based on traditional methods. Moreover, supported by machine learning and SVM technology, this study uses a DSP filter to convert the preprocessed EEG signal X into a time series, and adjusts the distance between the time series to classify the data. In order to solve the inconsistency of DSP algorithms, a multi-layer joint learning framework based on logistic regression model is proposed, and a brain-machine interface system of sports based on machine learning and SVM is constructed. In addition, this study designed a control experiment to improve the performance of the method proposed by this study. The research results show that the method in this paper has a certain practical effect and can be applied to sports.

Keywords: Machine learning, SVM, sports, brain-computer interface

DOI: 10.3233/JIFS-189481

Journal: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 4, pp. 6409-6420, 2021