sanjanadhiran16 / bci_knn_lda

Its a classification technique to detect emotion using brain waves and uses LDA and KNN .

BCI_KNN_LDA

Its a classification technique to detect emotion using brain waves and uses LDA and KNN .

Following is the way it was done in the code Step 1: Separate data from labels and create separate files for valence, arousal, dominance and liking. Step 2: divide labels into classes so as to detrmine emotions later on. Step 3: Split the data into training/testing sets Step 4: Apply feature scaleing, LDA and KNN classifier. Step 5: Apply Russell Circumplex model Step 6: Display emotions using emoticons

Russell Circumplex model: This model suggests that emotions are distributed in a two-dimensional circular space, containing arousal and valence dimensions. Arousal represents the vertical axis and valence represents the horizontal axis, while the center of the circle represents a neutral valence and a medium level of arousal.In this model, emotional states can be represented at any level of valence and arousal, or at a neutral level of one or both of these factors. Circumplex models have been used most commonly to test stimuli of emotion words, emotional facial expressions, and affective states.

DataBase used : Deap Database can be found at https://www.eecs.qmul.ac.uk/mmv/datasets/deap/ .