On the Utility of Power Spectral Techniques With Feature Selection Techniques for Effective Mental Task Classification in Noninvasive BCI

In this paper, classification of mental task-root brain-computer interfaces (BCIs) is being investigated. The tasks are dominant area investigations in BCI, which utmost interest as these system can be augmented life people having severe disabilities. performance BCI model primarily depends on the construction features from brain, electroencephalography (EEG), signal, and size feature vector, obtained through multiple channels. availability training samples to minimal for task classification. selection used increase ratio by getting rid irrelevant superfluous features. This paper suggests an approach augment a learning algorithm utility power spectral density (PSD) using selection. also deals comparative analysis multivariate univariate After applying above stated method, findings demonstrate substantial improvements Moreover, efficacy proposed endorsed carrying out robust ranking Friedman's statistical test finding best combinations compare various PSD methods.