• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International journal of swarm intelligence research >Particle Rider Optimization-Driven Classification for Brain-Computer Interface
【24h】

Particle Rider Optimization-Driven Classification for Brain-Computer Interface

机译：Particle Rider Optimization-Driven Classification for Brain-Computer Interface

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相关主题

摘要

The emerging technology for translating the intention of humans into control signals is the brain-computer interface (BCI). The BCI helps patients with complete motor dysfunction to interact with people. In this research, a method for abnormality assessment in humans from the perspective of the BCI was proposed by developing a hybrid optimization algorithm based on electroencephalography (EEG). The hybrid optimization algorithm, called particle rider optimization algorithm (PROA) is designed through the incorporation of particle swarm optimization (PSO) and rider optimization algorithm (ROA). The pre-processing is done for filtering the noise and removal of the artifact, in pre-processing, the noise is removed through the common average referencing (CAR) and Laplacian filters, whereas the artifacts are eliminated by principle component analysis (PCA). The features, such as Wavelet transform, statistical features, and spectral features are extracted for establishing the feature vector. Conversely, the statistical features, like mean, variance, kurtosis, skewness, entropy, and information gain are extracted. Similarly, the spectral features include, spectral flatness, spectral flux, and spectral kurtosis are also extracted. The classification is done using the RideNN in which the weights are trained using the developed PROA. The metrics, like sensitivity, accuracy, specificity, and kappa value are used for evaluating the performance of developed PROA-based RideNN. When compared to the existing BCI techniques, the proposed PROA-based RideNN scheme achieved a maximum accuracy, maximum sensitivity, maximum specificity, and maximum kappa value of 0.924, 0.936, 0.853, and 0.647, respectively.

著录项

获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号