Analysis of Emotion EEG Classification Based on GA-Fisher Classifier
Sheng Zhang, Jie Gao, Zhijie Chen · 2011
Emotion classification is a research hotspot in fields such as psychology and physiology. The categorical scales used recently need to be further researched for their subjective factors and accuracy influence. This paper presents an effective method which integrates GA-Fisher classifier and EEG, and we have got good classification effects by doing experiments on four emotions: excitement, fear, oscitancy, awaking. The results show that: (1) The classification accuracy rate between excitement and fear is 88.45%; and among excitement, fear and oscitancy for 86.71%; excitement, fear, oscitancy, and awaking for 84.70%; (2) the classifier introduced here is obviously better than the classical PCA-Fisher classifier (whose accuracy rate is 79.82% - 82.74%).