Improved fastICA to automatic noise removal for emotional classification
Maziyar Molavi · 2013
This paper explains the effect of denoising algorithm to classify emotional expression through Electroencephalogram (EEG). This research led to classify the EEG features due to emotions which was induced by the facial expression stimulus include of happy and sad and neutral cases. Time-frequency features was extracted to probe the ability of Improved Fast Independent Components Analysis (IFICA) based on optimization step size as a denoising mathematical tool which is used for data preprocessing. The features were reduced dimensionally by common spatial patterns (CSP). Support Vector Machine (SVM) was used to classify components which were evaluated for the effect of noise removal on data classification. The advantage ofIFICA was indicated by faster convergence and increasing the performance rate during the evaluation. Compare with the previous method, Fast Independent Component Analysis (FICA), the IFICA was significantly accurate during the emotional classification.