EEG Data Analytics to Distinguish Happy and Sad Emotions Based on Statistical Features

Affective computing is part of the important study of Human-Computer Interaction. Currently, EEG-based affective computing (emotion recognition) has become an interesting issue to be studied further. Emotions are not only closely related to aspects of HCI but also affect human health. Meanwhile, EEG is also considered a transparent tool in objectively revealing human emotions because the brain naturally produces EEG signals. This study focuses on comparing and classifying human emotions (happy and sad) based on EEG data. The channels used for recording EEG data are F7, F8, FP1, and FP2. Data preprocessing such as signal filtering, Independent Component Analysis, and Band Decomposition aims to clean the raw signal from artifacts and separate the signals according to specific frequency bands (Alpha, Beta, and Gamma). Then, statistical feature extraction is performed in the time domain to obtain the Mean values, Mean Absolute Value (MAV), and Standard Deviation values for further data analysis. The results show that emotion of happy has a higher feature value compared to emotion of sad. In the classification of happy and sad emotions using several algorithms, Random Forest signifies the highest classification accuracy (88.90%), compared to other algorithms such as SVM (86.70%), K-NN (88.87%), and Naive Bayes (86.63%).