Emotion Classification from EEG Brain Signal using Weighted Stacking of Ensemble Classifiers

Electroencephalogram (EEG) is very small amount of electrical signal generated in human brain due to the firing of neurons. Because of this signal information passes through our brain and a human being acts accordingly. These signals are very useful for analysis of various things that are happening in our brain. It can be used for any disease identification in brain, designing brain computer interface, emotion detection etc. In this paper we are going to consider the case of emotion detection by classification of EEG brain signals. For classification Machine Learning techniques are used. But alone one technique is not enough to provide results with high accuracy. Hence an ensemble of classifiers has been used by the help of weighted stacking approach. Through different experimental evaluations it has been proved that the proposed technique outperforms compared to simple machine learning classifier.