BP Neural Network-based Model for Evaluating User Interfaces of Human-computer Interaction System

Human-computer interaction system is the medium for human and computer. The rationality and intelligence of its design directly affect the work efficiency and execution ability of relevant practitioners. Traditional human-computer interaction evaluation usually adopts expert evaluation method. This method is difficult to evaluate objectively because of people’s subjective cognitive differences. Therefore, this paper proposes an intelligent evaluation method for complex human-computer interaction system based on BP neural network model. First, the known evaluation indicators are classified and organized, and five key evaluation indicators are optimized according to importance and relevance. Then the index is quantified into the evaluation function according to the fuzzy analytic hierarchy process. Finally, the data obtained by the simulation test is used as the training set and test set of the BP neural network, and then the evaluation model of the humancomputer interaction system is obtained.