EEG Recognition of Motor Imagery Based on SVM Ensemble
| Autor: | Jun Sheng, Jianze Liu, Jingyu Liu, Yangcheng Zhang, Jingwei Lv |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Computer Science::Machine Learning
Quantitative Biology::Neurons and Cognition medicine.diagnostic_test Computer science business.industry Pattern recognition Electroencephalography Support vector machine ComputingMethodologies_PATTERNRECOGNITION Motor imagery Robustness (computer science) Computer Science::Computer Vision and Pattern Recognition medicine Artificial intelligence business Classifier (UML) |
| Zdroj: | ICSAI |
| DOI: | 10.1109/icsai.2018.8599464 |
| Popis: | In Brain-Computer Interface (BCI) systems of motor imagery, a new electroencephalogram (EEG) identification method based on Ensemble Support Vector Machine(SVM) was proposed to solve the problem of low classification accuracy and weaker robustness for collecting EEGs during different time. Common Spatial Pattern (CSP) feature extraction algorithm was used and combined with Ensemble Support Vector Machine (SVM) as classifier for EEG. Besides, bagging and cross-validation ways were adopted in generation of Ensemble SVM. The experiment results showed the accuracy of Ensemble SVM was better than that of single SVM for different time collecting EEG and cross-validation way performed better than bagging method. Therefore, Ensemble SVM has stronger robustness and generalization ability compared to individual SVM, which develops a new idea for classifying EEG signals. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|