Research On Emotional Models Based On EEG Music

The brain is the human central nervous system, has played a leadership role in the body’s organs so, he is also on the human body’s most complex organs, research scientist never stopped. One important feature is that the brain can produce brain waves.Human activities neurons in the brain, resulting in the generation of EEG EEG, which combines the performance of the activity patterns of neurons, contains very rich information. In the field of brain research, most people can not ignore the study’s main research tool that is the EEG, EEG analysis as deepening understanding, related important research results are emerging. Brain wave EEG contains a wealth of emotional information, referred to recent emotion researchers focus of attention, but how to extract information from the emotional brain waves and a computer to make a prediction on emotion, becoming difficulties related research. This article is from the brain-computer interaction in this direction, this paper proposes a new way to identify computers emotion.Brain waves can have a musical and emotional reactions, the same reaction and emotional music, music can serve as EEG emotional medium. With modern machine learning methods, establish emotional classification model, computer EEG emotion and musical emotion common to make sense of the brain can make use of the form of music reflected, this is undoubtedly an important scientific value.New ideas presented in this paper is the medium of music study brain waves contain emotions, caught the general music and brain waves can be converted into common music presented this research new ideas. First, a standard MIDI music as a bridge to music in the form of brainwave signals emotional interpretation, designed to characterize the feature vector music emotion, establish emotional classification model, the establishment of machine learning sample database manually labeled. Finally, BP neural network algorithm, to build artificial neural network to classify EEG musical emotion, identification and analysis of EEG signal in the emotional and affective forecasting trends, combined with the known EEG emotional music validate the algorithm.