A WeChat Mini-program System with LSTM for The Emotional EEG Signal Recognition

As one of the advanced functions for human being, the emotion has a great influence on people's personality and mental health. EEG serves as a rapid measure method for neural signals that becomes an important way to evaluate different emotions. Some traditional machine learning techniques do not take into account the crucial temporal dynamic information in the EEG signals. However, with the recursive structure in time, the long and short time memory (LSTM) network in deep learning technology can solve this problem well. In this paper, a LSTM is designed and trained well to classify the emotional EEG, and then a WeChat mini-program system is constructed. The mini-program system incorporates with the LSTM to perform the EEG preprocessing, feature extraction, emotion classifying, and user management functions and so on. It can give feedback to the users about the emotional changes degree of pleasure and sobriety according to their EEG, which could serve as the emotion inspector as well as the entertainment tool for personal use.