Premium
Detecting the athlete's abnormal emotions before competition via support vector data description
Author(s) -
Gao Yue,
Ma Ziyin
Publication year - 2020
Publication title -
internet technology letters
Language(s) - English
Resource type - Journals
ISSN - 2476-1508
DOI - 10.1002/itl2.238
Subject(s) - support vector machine , electromyography , signal (programming language) , computer science , artificial intelligence , pattern recognition (psychology) , competition (biology) , affect (linguistics) , quality (philosophy) , wavelet , anomaly detection , speech recognition , data mining , psychology , machine learning , communication , neuroscience , ecology , philosophy , epistemology , biology , programming language
The athlete's emotions directly affect the quality of sport competition, which is an important factor that affects the action and competition effect. Therefore, it is very important to discover the athlete's abnormal emotions before competition in time. In this paper, we utilize the electromyography signal to monitor athlete emotions. First, we collect massive electromyography signal data from VICON system. Then, we remove the noises from collected electromyography signal data by using discrete wavelet transform. Lastly, we learn a support vector data description (SVDD) model from the clean data offline. The SVDD is a classical anomaly detection method which is widely used in finding abnormal events. The trained SVDD model is used to monitor the abnormal athlete's emotion automatically. The accuracy of monitoring athlete's abnormal emotions achieves 87.34%.