Emotion classification from.. - text for correction (proofreading) from user tpjung (original) (raw)

CorrectMyText Users

en ru de it fr es pt id uk tr

English language

The text to correct from user tpjung

Emotion classifica­tion from electroenc­ephalograp­hic (EEG) data has been a growing research topic in these years as its importance in potential applicatio­ns such as affective brain-computer interactio­ns and clinical assessment of mental illness. Many previous related works mainly focused on time-frequency features from the scalp channels, i. e. channel-based features, to obtain better classifica­tion accuracy. However, the far-field potentials creased by the brain activity in any area project widely across the scalp surface by passive volume conduction. Therefore, scalp EEG signals are the sum of activities from brain and non-brain sources, confoundin­g the scalp-level analysis. This study applies independen­t component analysis (ICA) to broadband spectral time series of scalp EEG to extract the brain sources that modulate the spectral contents of brain sources involved in emotional responses during music appreciati­on. Compared to the principal component analysis (PCA), the emotion-related independen­t spectral components provide informativ­e features which yield a remarkably high emotion-classifica­tion performanc­e and seem more neurophysi­ologically plausible in terms of dipolarity.

Language: English Language Skills: Native speaker, Proficiency

Login or register to correct this text!

You can correct this text if you set follow language skills of English in settings: Native speaker, Proficiency


Please correct texts in English:


Please help with translation: