Emotion recognition model construction method and system based on machine learning and physiological signals

The invention relates to an emotion recognition model construction method and system based on machine learning and physiological signals. The method comprises the following steps: acquiring various physiological data of a sample crowd, constructing a first sample data set, extracting each physiological signal feature, preprocessing to obtain a second sample data set, and dividing the second sample data set into a training set and a verification set according to a preset proportion; selecting a classification machine learning model to construct a primary model for each physiological signal, using the training set as input for training, and obtaining a plurality of emotion recognition models; analyzing the accuracy of each emotion recognition model, and calculating the correlation between different models and the stability of sample data, thereby obtaining a fusion weight of the emotion recognition model; and on the basis of the fusion weight, combining all emotion recognition models to construct a comprehensiv