An event classifier using EEG signals: An artificial neural network approach

An event classifier has been developed to analyze the collected EEG signals and distinguish between different events. The classifier introduced in this study was based on an artificial neural network model. The training process of the neural network required large amount of sample data and target results. A trained artificial neural network can then predict the outcome of an event based on the information of the corresponding EEG signal. The architecture of the artificial neural network involved hidden layers in addition to the input and output layers, which satisfied the non-linearity of the problem that the classifier was designed to solve. Experiments were conducted to validate this approach by using the classifier to distinguish whether subjects placed their fingers into hot or cold water. Validation results demonstrated the effectiveness of the classifier and its potential application in other fields.