At　present,　most　of　the　EEG　emotion　recognition　studies　have　taken　all　electric　shocks　or　filtered　electrodes　as　a　feature　and　they　are　integrated　(combined)　with　simple　features　that　are　extracted　from　other　signals　as　a　single　classifier　Emotional　classification,　but　there　are　problems　such　as　low　efficiency　and　low　accuracy.　Aiming　at　this　problem,　this　paper　proposes　an　EEG　emotion　classification　method　based　on　AdaBoost　classifier　to　optimize　the　algorithm,　dividing　the　feature　samples　into　different　channels,　and　then　using　the　genetic　algorithm　to　identify　the　EEG　emotion,　and　the　algorithm　achieves　in　Matlab.　The　experimental　result　shows　that　the　average　recognition　rate　of　EEG　is　90.8%,　which　is　superior　to　the　current　single　classifier　and　majority　voting　method,　and　it　has　better　stability　and　generalization　performance.