Title :
Extracting Impact Characteristics of Sports Training on EEG by Genetic Algorithm
Author :
Li, Jing ; Wang, Wei
Author_Institution :
Inst. of Educ. Sci., Nanjing Normal Univ., Nanjing, China
Abstract :
It has been proved that it´s helpful to improve brain electrical activity mode through effective exercise training. According to Gardner´s multiple intelligences theory, after a long period of professional training, college students of different specialties may also have intellectual independence and EEG specificity. In this study, the author chose motor imagery EEG of college students specialized in mathematical logic and sports education as the experimental data, and used Genetic Algorithm and BP neural network for EEG feature selection and pattern classification, in order to find impact characteristics of long-term exercise training on EEG. The experiment results showed there were significantly difference between these two specialties on the change of α and β band power in the left prefrontal area, the right parietal, the left middle temporal lobe and the right middle temporal lobe in the process of motor imagery tasks. Among these, the left and the right middle temporal lobe and he wave have the greatest impact.
Keywords :
bioelectric phenomena; electroencephalography; feature extraction; genetic algorithms; medical signal processing; neural nets; pattern classification; signal classification; sport; training; α band power; β band power; BP neural network; EEG feature selection; Gardner´s multiple intelligence theory; brain electrical activity mode; college students; exercise training; genetic algorithm; impact characteristic extraction; intellectual independence; mathematical logic; middle temporal lobe; motor imagery EEG; parietal; pattern classification; prefrontal area; professional training; sport education; Accuracy; Biological neural networks; Electrodes; Electroencephalography; Genetic algorithms; Temporal lobe; Training; BP neural network; EEG; GA; Multiple intelligences; Sports Training;
Conference_Titel :
Complexity and Data Mining (IWCDM), 2011 First International Workshop on
Conference_Location :
Nanjing, Jiangsu
Print_ISBN :
978-1-4577-2007-9
DOI :
10.1109/IWCDM.2011.48