An Approach For Finding Emotions Using Seed Dataset With Knn Classifier

Abstract

Emotions are an indispensable component of our daily lives. Nonetheless, brain-computer interface (BCI) systems have not achieved the requisite level to interpret emotions. Programmed feeling acknowledgment based on BCI frameworks has been a point of extraordinary inquiry within the final few decades. Electroencephalogram (EEG) signals are one of the significant assets for these frameworks EEG may be a physiological flag recorded from brain work out within the frame of brain waves through the scalp. The most advantage of utilizing EEG signals is that it reflects the genuine feeling and can effectively be prepared by computer frameworks. A dummy dataset can be used and filled with EEG data to compute and categorize these signals generated from EEG signals. The dataset used here is Seed, and it can be accustomed by a machine learning technique called the K-nearest neighborhood (KNN) algorithm to systematize the data. Experimental performance achieved through the categorizing values of 94.06% during the classification in the Seed. This proposed method shows that emotion recognition like positive, neutral, and negative is possible through EEG signals.